- Email: [email protected]
Upper gastrointestinal safety evaluation of parecoxib sodium, a new parenteral cyclooxygenase-2—specific inhibitor, compared with ketorolac, naproxen, and placebo
CLINICAL THERAPEUTICS®/VOL.23, NO. 9, 2001
Upper Gastrointestinal Safety Evaluation of Parecoxib Sodium, a New Parenteral Cyclooxygenase-2-Specific Inhibitor, Compared with Ketorolac, Naproxen, and Placebo Stuart L Harris, MD, PhD, 1 Michael Kuss, BS, 2 Richard C. Hubbard, MD, 2 and Jay L. Goldstein, MD s 1SeaView Research, Miami, Florida, 2Pharmacia Research and Development, Skokie, and 3University of Illinois at Chicago, Chicago, Illinois
ABSTRACT
Background: Conventional nonsteroidal anti-inflammatory drugs (NSAIDs) are associated with an increased risk of ulcers and upper gastrointestinal (GI) ulcer complications, which has been attributed to the inhibition of cyclooxygenase-1. These risks are usually increased in elderly populations. Parecoxib sodium is an injectable prodrug of the cyclooxygenase-2-specific inhibitor valdecoxib that has exhibited analgesic activity in previous trials. Objective: The purpose of this study was to compare the GI safety and tolerability profile of parecoxib sodium with that of ketorolac, naproxen, and placebo in a 7-day endoscopic trial in elderly subjects. Methods: This was a randomized, double-blind, double-dummy, placebo-controlled, parallel-group study. After a normal baseline endoscopy, healthy elderly subjects aged 66 to 75 years were randomized to receive IV parecoxib sodium (10 mg BID), oral naproxen (500 mg BID), or placebo for 7 days, or placebo for 2 days followed by IV ketorolac (15 mg QID) for 5 days. Endoscopy was performed again after 7 days. Results: Among the first 17 subjects enrolled, ulcers were observed in all treatment groups except the parecoxib sodium group (ketorolac, 4/4 subjects; naproxen, 2/4 subjects; and placebo, 2/5 subjects). Four subjects in the ketorolac group and 1 subject in the naproxen group had multiple gastric ulcers or combined gastric and duodenal ulcers. Because of the unexpectedly high incidence of gastroduodenal ulcers observed, the study was terminated early and the randomization blind broken. Conclusion: These findings suggest that elderly patients may be at risk for GI ulceration even after short-term use of the conventional NSAIDs ketorolac and naproxen. Key words: parecoxib sodium, ketorolac, naproxen, nonsteroidal anti-inflammatory drugs, gastrointestinal toxicity, COX-2 inhibitor. (Clin Ther. 2001 ;23:1422-1428) Accepted for publication June 27, 2001. Printed in the USA. Reproduction in whole or part is not permitted.
1422
014%2918/01/$ ! 9.00
S.I. HARRIS ET AL.
INTRODUCTION Nonselective, conventional nonsteroidal anti-inflammatory drugs (NSAIDs) are widely used for the treatment of arthritis symptoms and for analgesia but are associated with substantial dose-dependent gastrointestinal (GI) toxicity. Epidemiologic studies estimate that the risk of GI complications and death is 3 to 10 times higher in patients who take conventional NSAIDs regularly than in those who do notl; other studies have also shown an increased risk of symptomatic ulcer disease and upper gastrointestinal (UGI) bleeding with NSAID use. Elderly subjects (ie, those aged >65 years) are more likely to use NSAIDs for osteoarthritis and are at higher risk for ulcer complications.2 5 The increased GI toxicity associated with NSAID use has been attributed to the inhibition of cyclooxygenase (COX) and the subsequent reduced production of prostaglandins that help maintain the integrity of the gastric mucosal barrier. COX is present in 2 forms in human cells, a constitutive form (COX-l) widely expressed in nearly all tissues throughout the body including the UGI tract, and an inducible form (COX-2) predominantly expressed in inflamed tissues. 6,7 Conventional NSAIDs, such as ketorolac and naproxen, exert anti-inflammatory, antipyretic, and analgesic effects through inhibition of COX-2. 8,9 Inhibition of gastric COX-1 by NSAIDs is thought to predispose patients to GI mucosal injury, such as erosions and ulcers and UGI ulcer complications.l°,l i Parecoxib sodium is an injectable prodrug formulation of the COX-2-specific inhibitor valdecoxib 12 that has demonstrated significant analgesic activity in a
postsurgical dental pain model. 13 Compared with conventional NSAIDs, which inhibit both COX-1 and COX-2, COX-2specific inhibitors have demonstrated a reduced risk of GI ulceration and ulcer complications. 3,5 Thus, it was hypothesized that parecoxib sodium might cause less GI damage than conventional NSAIDs, a property that would be of benefit in the elderly population, in which nonselective NSAID toxicity is heightened. The purpose of this study was to compare the effects of parecoxib sodium, ketorolac, naproxen, and placebo on the UGI mucosa of healthy elderly subjects.
SUBJECTS AND METHODS This was a randomized, double-blind, double-dummy, placebo-controlled parallelgroup study comparing the GI toxicity of parecoxib sodium, naproxen, and ketorolac. Naproxen was chosen as one of the positive controls because there is considerable prior experience with this agent in this type of study and it is a widely used oral analgesic. As such, use of naproxen as a comparator has clinical relevance. This study was approved by the institutional review board at the participating center, and written informed consent was obtained from each subject. Healthy older adult subjects aged 66 to 75 years with endoscopically confirmed, normal UGI mucosa and no clinically significant findings on physical examination or laboratory testing were enrolled. Patients with a history of UGI ulcers, bleeding, surgery, or evidence of petechiae, erosions, ulcerations, or active bleeding at the pretreatment endoscopy were excluded. Use of NSAIDs (including aspirin) or antiulcer medications (including H 2 antagonists and proton pump inhibitors) was 1423
CLINICAL THERAPEUTICS ®
not permitted in the 2 weeks before randomization or during the course of the study. Subjects were admitted to a closed clinical research facility and received IV parecoxib sodium (10 mg BID), oral naproxen (500 mg BID), or placebo (0.9% sodium chloride) for 7 days, or placebo for 2 days followed by IV ketorolac (15 mg QID) for 5 days. This dosage regimen for ketorolac was consistent with dosage recommendations for elderly patients, and the dosing restrictions for ketorolac (US Summary of Product Labeling, August 1997). 14 Dosing was conducted in a double-dummy design so that each patient received both oral and IV medication. Subjects remained in the facility for the duration of the study and received a normal diet free of alcohol, fried foods, and heavily spiced foods. Helicobacter pylori serologic testing was performed before enrollment using the HexSure test (Beckman Coulter, Inc, Fullerton, California). Posttreatment endoscopy was performed within 24 hours of the final dose, and H pylori testing was repeated using the CLO test (Delta West, Pty, Ltd, Bentley, Western Australia). Ulcers were defined as lesions >3 mm in diameter with appreciable depth. 15 The occurrence of adverse events was monitored throughout the study. This trial was initially designed to enroll 160 subjects (40 per treatment group). This sample size was determined to compare the gastroduodenal ulcer rates of 0.001% with >18.8% using a 2-sided test, with c~ = 0.05 and 80% power. These rates were obtained from a previous similar study of a specific COX-2 inhibitor and naproxen in healthy subjects. 15 However, after reviewing the blinded data for 17 enrolled subjects, both the investigator and sponsor concurred that the study should be terminated early due to a high incidence of endoscopic UGI ulcers. Statisti1424
cal tests were therefore not performed because of the small sample size.
RESULTS At the time of early termination, 8 men and 9 women had enrolled in the trial. Treatment groups were similar with respect to sex and mean age (70 years in the placebo, parecoxib sodium, and ketorolac groups; 72 years in the naproxen group). The number and size of ulcers observed after 7 days are shown in the table. All ulcers were detected endoscopically at the completion of study drug administration. Ulcers were observed in all treatment groups except the parecoxib sodium group. None of the ulcers were noted to have bleeding stigmata at the time of endoscopy. In the ketorolac and naproxen treatment groups, several patients had multiple gastric or combined gastric and duodenal ulcers. Because of the high overall incidence of ulcers, and the number of patients with multiple ulcers, the study was terminated early at the request of the investigator, and the randomization blind was broken. Mild adverse events were reported by 4 of the 17 subjects. Two subjects in the placebo group experienced headache, peripheral pain, and anemia; 1 subject in the parecoxib sodium group experienced toothache; and 1 subject in the ketorolac group experienced diarrhea and inflammation at the injection site. The subject with anemia in the placebo group also had an increased level of eosinophils (>10%) and a decreased level of neutrophils (<30%) at the posttreatment assessment, which were considered to be clinically relevant. However, both values had returned to the normal range at a follow-up visit 4 days later. Pretreatment to post-
S.I. HARRIS ET AL.
Table. Characteristics of gastroduodenal ulcers in elderly subjects treated with intravenous ketorolac (n = 4) for 5 days, or placebo (n = 5), oral naproxen (n = 4), or intravenous parecoxib sodium (n = 4) for 7 days. Subject
Drug and Dose
No. of Ulcers
0002
Ketorolac 15 mg QID
5
Gastric 3 @ 0.3 cm Gastric 1 @ 0.5 cm Gastric 1 @ 1.0 cm
0011
Ketorolac 15 mg QID
5
Gastric 3 @ 0.5 cm Gastric 1 @ 0.6 cm Gastric 1 @ 0.8 cm
0013
Ketorolac 15 mg QID
2
Gastric 2 @ 0.5 cm
0203
Ketorolac 15 mg QID
10
0010
Placebo
1
Gastric 1 @ 1.0 cm
0201
Placebo
1
Gastric 1 @ 1.0 cm
0012
Naproxen 500 mg BID
1
Gastric 1 @ 0.8 cm
0202
Naproxen 500 mg BID
5
Gastric 3 @ 0.3 cm Gastric 1 @ 0.5 cm Gastric 1 @ 0.6 cm
treatment changes in mean laboratory values were clinically insignificant (based on prospectively defined variations relative to the normal range or baseline values), and no other posttreatment values exceeded normal ranges. There was no clear relationship between H pylori status and ulcer development because o f the small sample size. Of the 2 subjects who developed ulcers in the placebo group, 1 was H pylori-positive and 1 was H pylorinegative by serology testing, and both subjects were negative by the CLO test. Three o f the 4 subjects in the ketorolac treatment group who developed ulcers tested positive for H pylori on serology testing, and 2 of the 4 tested positive on the C L O test. In the naproxen treatment group, 1 of the 2 subjects with ulcers tested positive
Ulcer Type and Size
Gastric 7 @ 0.5 cm Duodenal 3 @ 0.5 cm
on serology testing, whereas both tested negative on the CLO test. DISCUSSION Case reports and retrospective studies of GI ulceration and/or ulcer complications following injectable treatment with ketorolac have been reported previously. 16-2° However, there have been few systematic, prospective, blinded endoscopic trials evaluating the UGI toxicity of ketorolac. 2k22 A key finding of this study was the unexpectedly high incidence of frank gastroduodenal ulceration in elderly subjects receiving ketorolac at standard acute analgesia doses (15 mg IV QID). Although the sample size in this study is too small to make definitive conclusions, the obser-
1425
CLINICAL THERAPEUTICS*
vation that all 4 subjects in the ketorolac treatment group developed multiple ulcers may suggest that the ulcerogenicity of ketorolac in the elderly, even after only short-term use at the recommended dose, may be greater than previously suggested by episodic case reports. 16-20 Reasons for the high ulceration rate observed in this study, other than the ulcerogenic effects of NSAIDs, are unclear. Misidentification of tablets was ruled out by careful retrospective examination. All ulcers were evaluated by an experienced endoscopist and were >3 mm in diameter with perceptible depth and thus were not erosions. The higher rate in the placebo arm might be attributed to the older poputation enrolled in this study. However, the small sample size precludes any conclusions from being drawn. Our results parallel the findings of a retrospective case-control study in which the authors estimated that ketorolac increased the risk of a peptic ulcer, with an odds ratio of 9.8 (95% CI, 3.4-28.1). 16 However, the results and generalizability of the findings of this retrospective study are limited because the dose and duration of ketorolac use were not reported. Although there have been several case reports of ketorolac-associated ulcer complications, the ketorolac dosage in these studies often exceeded the recommended dose for the elderly (15 mg IV). 17-19 In contrast to the ketorolac group, there were no ulcers observed in the parecoxib sodium group. In the naproxen arm, 2 of 4 subjects had ulcers and 1 of them had 5 ulcers. This ulcerogenic effect of naproxen is supported by other 7-day endoscopic trials. 15,23 One of these studies reported a 19% incidence of gastric ulcers in naproxen-treated subjects, and the other found that naproxen was associated with 1426
significantly worse endoscopy scores compared with placebo (mean score 1.92 and 0.25, respectively). These brief results suggest the need for vigilance when prescribing injectable nonspecific COX inhibitors to an elderly population. CONCLUSION These findings suggest that elderly patients may be at risk for GI ulceration even after short-term use of the conventional NSAIDs ketorolac and naproxen.
ACKNOWLEDGMENT This study was supported by Pharmacia Research and Development, Skokie, Illinois.
REFERENCES 1. Hawkey CJ, Karrasch JA, Szczepanski L, et al, for the Omeprazole Versus Misoprostol for NSAID-Induced Ulcer Management (OMNIUM) Study Group. Omeprazole compared with misoprostol for ulcers associated with nonsteroidal antiinflammatory drugs. N Engl J Mecl. 1998; 338:727-734. 2. Griffin MR. Epidemiology of nonsteroidal anti-inflammatory drug-associated gastrointestinal injury. Am J Med. 1998;104: 23S-29S. 3. Silverstein FE, Faich (3, (3oldstein JL, et al. Gastrointestinal toxicity with celecoxih vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis. The CLASS study: A randomized controlled clinical trial. Celecoxib Long-Term Arthritis Safety Study. JAMA. 2000;284: 1247-1255. 4. Silverstein FE, Graham DY, Senior JR, et al. Misoprostol reduces serious gastrointestinal complications in patients with
S.I. HARRIS ET AL.
rheumatoid arthritis receiving nonsteroidal anti-inflammatory drugs. A randomized, double-blind, placebo-controlled trial. Ann Intern Med. 1995;123:241-249.
parecoxib sodium: A potent and selective inhibitor of COX-2 for parenteral administration. J M e d Chem. 2000;43:1661-1663.
5. Goldstein JL, Silverstein FE, Agrawal NM, et al. Reduced risk of upper gastrointestinal ulcer complications with celecoxib, a novel COX-2 inhibitor. A m J Gastroenterol. 2000;95:1681-1690.
13. Kuss M, Mehlisch D, Bauman A, et al. Analgesic activity of single IV doses of parecoxib and ketorolac in postoperative dental pain. Poster presented at: 9th World Congress on Pain (WCP); August 22-27, 1999; Vienna, Austria; Board No. 249.
6. Lipsky PE, Abramson SB, Crofford L, et al. The classification of cyclooxygenase inhibitors. J Rheumatol. 1998;25(Suppl 12): 2298-2303.
14. Toradol® (ketorolac tromethamine). In: Physicians' Desk Reference ®. Montvale, NJ: Medical Economics Co; 2001:2789-2792.
7. Seibert K, Zhang Y, Leahy K, et al. Pharmacological and biochemical demonstration of the role of cyclooxygenase-2 in inflammation and pain. Proc Natl Acad Sci U S A. 1994;9l:12013-12017. 8. Vane JR. Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nat New Biol. 1971; 231:232-235. 9. Davies P, Bailey PJ, Goldenburg MM, Ford-Hutchinson AW. The role of arachidonic acid oxygenation products in pain and inflammation. Annu Rev lmmunol. 1984;2:335-357. 10. Garcia Rodriguez LA, Jick H. Risk of upper gastrointestinal bleeding and perforation associated with individual nonsteroidal anti-inflammatory drugs. Lancet. 1994;343:769-772. 11. Dubois RN, Abramson SB, Crofford L, et al. Cyclooxygenase in biology and disease. FASEB J. 1998;12:1063-1073. 12. Talley JJ, Bertenshaw SR, Brown DL, et al. N-[[(5-Methyl-3-phenylisoxazol-4-yl)phenyl]sulfonyl]propanamide, sodium salt,
15. Simon LS, Lanza FL, Lipsky PE, et al. Preliminary study of the safety and efficacy of SC-58635, a novel cyclooxygenase 2 inhibitor: Efficacy and safety in two placebo-controlled trials in osteoathritis and rheumatoid arthritis, and studies of gastrointestinal and platelet effects. Arthritis Rheum. 1998;41:1591-1602. 16. Traversa G, Walker AM, Ippolito FM, et al. Gastroduodenal toxicity of different nonsteroidal antiinflammatory drugs. Epidemiology. 1995;6:49-54. 17. Maliekal J, Elboim CM. Gastrointestinal complications associated with intramuscular ketorolac tromethamine therapy in the elderly. Ann Pharmacother. 1995;29: 698-701. 18. Wolfe PA, Polhamus CD, Kubik C, et al. Giant duodenal ulcers associated with the postoperative use of ketorolac: Report of three cases. Am J Gastroenterol. 1994;89: 1110-1111. 19. Estes LL, Fuhs DW, Heaton AH, et al. Gastric ulcer perforation associated with the use of injectable ketorolac. Ann Pharmacother. 1993;27:42-43.
1427
CLINICAL THERAPEUTICS®
20. Quigley EM, Donovan JP, Livingston WC. Ketorolac-related giant gastric ulcers. Am J Gastroenterol. 1994;89:631-632. 21. Lanza FL, Karlin DA, Yee JE Doubleblind placebo controlled endoscopic study comparing the mucosal injury seen with an orally and parenterally administered new nonsteroidal analgesic ketorolac tromethamine at therapeutic and supratherapeutic doses. Am J GastroenteroL 1987; 82:939a.
22. Nuutinen LS, Laitinen JO, Salomaki TE. A risk-benefit appraisal of injectable NSAIDs in the management of postoperative pain. Drug Saf 1993;9:380-393. 23. Lanza E Rack ME Lynn M, et al. An endoscopic comparison of the effects of etodolac, indomethacin, ibuprofen, naproxen, and placebo on the gastrointestinal mucosa. J Rheumatol. 1987;14" 338-341.
Address correspondence to: Jay L. Goldstein, MD, Professor of Medicine, University ot Illinois at Chicago, 840 South Wood Street (m/c 787), Room 734 CSB, Chicago, IL 60612. E-mail: [email protected]
1428
Upper Gastrointestinal Safety Evaluation of Parecoxib Sodium, a New Parenteral Cyclooxygenase-2-Specific Inhibitor, Compared with Ketorolac, Naproxen, and Placebo Stuart L Harris, MD, PhD, 1 Michael Kuss, BS, 2 Richard C. Hubbard, MD, 2 and Jay L. Goldstein, MD s 1SeaView Research, Miami, Florida, 2Pharmacia Research and Development, Skokie, and 3University of Illinois at Chicago, Chicago, Illinois
ABSTRACT
Background: Conventional nonsteroidal anti-inflammatory drugs (NSAIDs) are associated with an increased risk of ulcers and upper gastrointestinal (GI) ulcer complications, which has been attributed to the inhibition of cyclooxygenase-1. These risks are usually increased in elderly populations. Parecoxib sodium is an injectable prodrug of the cyclooxygenase-2-specific inhibitor valdecoxib that has exhibited analgesic activity in previous trials. Objective: The purpose of this study was to compare the GI safety and tolerability profile of parecoxib sodium with that of ketorolac, naproxen, and placebo in a 7-day endoscopic trial in elderly subjects. Methods: This was a randomized, double-blind, double-dummy, placebo-controlled, parallel-group study. After a normal baseline endoscopy, healthy elderly subjects aged 66 to 75 years were randomized to receive IV parecoxib sodium (10 mg BID), oral naproxen (500 mg BID), or placebo for 7 days, or placebo for 2 days followed by IV ketorolac (15 mg QID) for 5 days. Endoscopy was performed again after 7 days. Results: Among the first 17 subjects enrolled, ulcers were observed in all treatment groups except the parecoxib sodium group (ketorolac, 4/4 subjects; naproxen, 2/4 subjects; and placebo, 2/5 subjects). Four subjects in the ketorolac group and 1 subject in the naproxen group had multiple gastric ulcers or combined gastric and duodenal ulcers. Because of the unexpectedly high incidence of gastroduodenal ulcers observed, the study was terminated early and the randomization blind broken. Conclusion: These findings suggest that elderly patients may be at risk for GI ulceration even after short-term use of the conventional NSAIDs ketorolac and naproxen. Key words: parecoxib sodium, ketorolac, naproxen, nonsteroidal anti-inflammatory drugs, gastrointestinal toxicity, COX-2 inhibitor. (Clin Ther. 2001 ;23:1422-1428) Accepted for publication June 27, 2001. Printed in the USA. Reproduction in whole or part is not permitted.
1422
014%2918/01/$ ! 9.00
S.I. HARRIS ET AL.
INTRODUCTION Nonselective, conventional nonsteroidal anti-inflammatory drugs (NSAIDs) are widely used for the treatment of arthritis symptoms and for analgesia but are associated with substantial dose-dependent gastrointestinal (GI) toxicity. Epidemiologic studies estimate that the risk of GI complications and death is 3 to 10 times higher in patients who take conventional NSAIDs regularly than in those who do notl; other studies have also shown an increased risk of symptomatic ulcer disease and upper gastrointestinal (UGI) bleeding with NSAID use. Elderly subjects (ie, those aged >65 years) are more likely to use NSAIDs for osteoarthritis and are at higher risk for ulcer complications.2 5 The increased GI toxicity associated with NSAID use has been attributed to the inhibition of cyclooxygenase (COX) and the subsequent reduced production of prostaglandins that help maintain the integrity of the gastric mucosal barrier. COX is present in 2 forms in human cells, a constitutive form (COX-l) widely expressed in nearly all tissues throughout the body including the UGI tract, and an inducible form (COX-2) predominantly expressed in inflamed tissues. 6,7 Conventional NSAIDs, such as ketorolac and naproxen, exert anti-inflammatory, antipyretic, and analgesic effects through inhibition of COX-2. 8,9 Inhibition of gastric COX-1 by NSAIDs is thought to predispose patients to GI mucosal injury, such as erosions and ulcers and UGI ulcer complications.l°,l i Parecoxib sodium is an injectable prodrug formulation of the COX-2-specific inhibitor valdecoxib 12 that has demonstrated significant analgesic activity in a
postsurgical dental pain model. 13 Compared with conventional NSAIDs, which inhibit both COX-1 and COX-2, COX-2specific inhibitors have demonstrated a reduced risk of GI ulceration and ulcer complications. 3,5 Thus, it was hypothesized that parecoxib sodium might cause less GI damage than conventional NSAIDs, a property that would be of benefit in the elderly population, in which nonselective NSAID toxicity is heightened. The purpose of this study was to compare the effects of parecoxib sodium, ketorolac, naproxen, and placebo on the UGI mucosa of healthy elderly subjects.
SUBJECTS AND METHODS This was a randomized, double-blind, double-dummy, placebo-controlled parallelgroup study comparing the GI toxicity of parecoxib sodium, naproxen, and ketorolac. Naproxen was chosen as one of the positive controls because there is considerable prior experience with this agent in this type of study and it is a widely used oral analgesic. As such, use of naproxen as a comparator has clinical relevance. This study was approved by the institutional review board at the participating center, and written informed consent was obtained from each subject. Healthy older adult subjects aged 66 to 75 years with endoscopically confirmed, normal UGI mucosa and no clinically significant findings on physical examination or laboratory testing were enrolled. Patients with a history of UGI ulcers, bleeding, surgery, or evidence of petechiae, erosions, ulcerations, or active bleeding at the pretreatment endoscopy were excluded. Use of NSAIDs (including aspirin) or antiulcer medications (including H 2 antagonists and proton pump inhibitors) was 1423
CLINICAL THERAPEUTICS ®
not permitted in the 2 weeks before randomization or during the course of the study. Subjects were admitted to a closed clinical research facility and received IV parecoxib sodium (10 mg BID), oral naproxen (500 mg BID), or placebo (0.9% sodium chloride) for 7 days, or placebo for 2 days followed by IV ketorolac (15 mg QID) for 5 days. This dosage regimen for ketorolac was consistent with dosage recommendations for elderly patients, and the dosing restrictions for ketorolac (US Summary of Product Labeling, August 1997). 14 Dosing was conducted in a double-dummy design so that each patient received both oral and IV medication. Subjects remained in the facility for the duration of the study and received a normal diet free of alcohol, fried foods, and heavily spiced foods. Helicobacter pylori serologic testing was performed before enrollment using the HexSure test (Beckman Coulter, Inc, Fullerton, California). Posttreatment endoscopy was performed within 24 hours of the final dose, and H pylori testing was repeated using the CLO test (Delta West, Pty, Ltd, Bentley, Western Australia). Ulcers were defined as lesions >3 mm in diameter with appreciable depth. 15 The occurrence of adverse events was monitored throughout the study. This trial was initially designed to enroll 160 subjects (40 per treatment group). This sample size was determined to compare the gastroduodenal ulcer rates of 0.001% with >18.8% using a 2-sided test, with c~ = 0.05 and 80% power. These rates were obtained from a previous similar study of a specific COX-2 inhibitor and naproxen in healthy subjects. 15 However, after reviewing the blinded data for 17 enrolled subjects, both the investigator and sponsor concurred that the study should be terminated early due to a high incidence of endoscopic UGI ulcers. Statisti1424
cal tests were therefore not performed because of the small sample size.
RESULTS At the time of early termination, 8 men and 9 women had enrolled in the trial. Treatment groups were similar with respect to sex and mean age (70 years in the placebo, parecoxib sodium, and ketorolac groups; 72 years in the naproxen group). The number and size of ulcers observed after 7 days are shown in the table. All ulcers were detected endoscopically at the completion of study drug administration. Ulcers were observed in all treatment groups except the parecoxib sodium group. None of the ulcers were noted to have bleeding stigmata at the time of endoscopy. In the ketorolac and naproxen treatment groups, several patients had multiple gastric or combined gastric and duodenal ulcers. Because of the high overall incidence of ulcers, and the number of patients with multiple ulcers, the study was terminated early at the request of the investigator, and the randomization blind was broken. Mild adverse events were reported by 4 of the 17 subjects. Two subjects in the placebo group experienced headache, peripheral pain, and anemia; 1 subject in the parecoxib sodium group experienced toothache; and 1 subject in the ketorolac group experienced diarrhea and inflammation at the injection site. The subject with anemia in the placebo group also had an increased level of eosinophils (>10%) and a decreased level of neutrophils (<30%) at the posttreatment assessment, which were considered to be clinically relevant. However, both values had returned to the normal range at a follow-up visit 4 days later. Pretreatment to post-
S.I. HARRIS ET AL.
Table. Characteristics of gastroduodenal ulcers in elderly subjects treated with intravenous ketorolac (n = 4) for 5 days, or placebo (n = 5), oral naproxen (n = 4), or intravenous parecoxib sodium (n = 4) for 7 days. Subject
Drug and Dose
No. of Ulcers
0002
Ketorolac 15 mg QID
5
Gastric 3 @ 0.3 cm Gastric 1 @ 0.5 cm Gastric 1 @ 1.0 cm
0011
Ketorolac 15 mg QID
5
Gastric 3 @ 0.5 cm Gastric 1 @ 0.6 cm Gastric 1 @ 0.8 cm
0013
Ketorolac 15 mg QID
2
Gastric 2 @ 0.5 cm
0203
Ketorolac 15 mg QID
10
0010
Placebo
1
Gastric 1 @ 1.0 cm
0201
Placebo
1
Gastric 1 @ 1.0 cm
0012
Naproxen 500 mg BID
1
Gastric 1 @ 0.8 cm
0202
Naproxen 500 mg BID
5
Gastric 3 @ 0.3 cm Gastric 1 @ 0.5 cm Gastric 1 @ 0.6 cm
treatment changes in mean laboratory values were clinically insignificant (based on prospectively defined variations relative to the normal range or baseline values), and no other posttreatment values exceeded normal ranges. There was no clear relationship between H pylori status and ulcer development because o f the small sample size. Of the 2 subjects who developed ulcers in the placebo group, 1 was H pylori-positive and 1 was H pylorinegative by serology testing, and both subjects were negative by the CLO test. Three o f the 4 subjects in the ketorolac treatment group who developed ulcers tested positive for H pylori on serology testing, and 2 of the 4 tested positive on the C L O test. In the naproxen treatment group, 1 of the 2 subjects with ulcers tested positive
Ulcer Type and Size
Gastric 7 @ 0.5 cm Duodenal 3 @ 0.5 cm
on serology testing, whereas both tested negative on the CLO test. DISCUSSION Case reports and retrospective studies of GI ulceration and/or ulcer complications following injectable treatment with ketorolac have been reported previously. 16-2° However, there have been few systematic, prospective, blinded endoscopic trials evaluating the UGI toxicity of ketorolac. 2k22 A key finding of this study was the unexpectedly high incidence of frank gastroduodenal ulceration in elderly subjects receiving ketorolac at standard acute analgesia doses (15 mg IV QID). Although the sample size in this study is too small to make definitive conclusions, the obser-
1425
CLINICAL THERAPEUTICS*
vation that all 4 subjects in the ketorolac treatment group developed multiple ulcers may suggest that the ulcerogenicity of ketorolac in the elderly, even after only short-term use at the recommended dose, may be greater than previously suggested by episodic case reports. 16-20 Reasons for the high ulceration rate observed in this study, other than the ulcerogenic effects of NSAIDs, are unclear. Misidentification of tablets was ruled out by careful retrospective examination. All ulcers were evaluated by an experienced endoscopist and were >3 mm in diameter with perceptible depth and thus were not erosions. The higher rate in the placebo arm might be attributed to the older poputation enrolled in this study. However, the small sample size precludes any conclusions from being drawn. Our results parallel the findings of a retrospective case-control study in which the authors estimated that ketorolac increased the risk of a peptic ulcer, with an odds ratio of 9.8 (95% CI, 3.4-28.1). 16 However, the results and generalizability of the findings of this retrospective study are limited because the dose and duration of ketorolac use were not reported. Although there have been several case reports of ketorolac-associated ulcer complications, the ketorolac dosage in these studies often exceeded the recommended dose for the elderly (15 mg IV). 17-19 In contrast to the ketorolac group, there were no ulcers observed in the parecoxib sodium group. In the naproxen arm, 2 of 4 subjects had ulcers and 1 of them had 5 ulcers. This ulcerogenic effect of naproxen is supported by other 7-day endoscopic trials. 15,23 One of these studies reported a 19% incidence of gastric ulcers in naproxen-treated subjects, and the other found that naproxen was associated with 1426
significantly worse endoscopy scores compared with placebo (mean score 1.92 and 0.25, respectively). These brief results suggest the need for vigilance when prescribing injectable nonspecific COX inhibitors to an elderly population. CONCLUSION These findings suggest that elderly patients may be at risk for GI ulceration even after short-term use of the conventional NSAIDs ketorolac and naproxen.
ACKNOWLEDGMENT This study was supported by Pharmacia Research and Development, Skokie, Illinois.
REFERENCES 1. Hawkey CJ, Karrasch JA, Szczepanski L, et al, for the Omeprazole Versus Misoprostol for NSAID-Induced Ulcer Management (OMNIUM) Study Group. Omeprazole compared with misoprostol for ulcers associated with nonsteroidal antiinflammatory drugs. N Engl J Mecl. 1998; 338:727-734. 2. Griffin MR. Epidemiology of nonsteroidal anti-inflammatory drug-associated gastrointestinal injury. Am J Med. 1998;104: 23S-29S. 3. Silverstein FE, Faich (3, (3oldstein JL, et al. Gastrointestinal toxicity with celecoxih vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis. The CLASS study: A randomized controlled clinical trial. Celecoxib Long-Term Arthritis Safety Study. JAMA. 2000;284: 1247-1255. 4. Silverstein FE, Graham DY, Senior JR, et al. Misoprostol reduces serious gastrointestinal complications in patients with
S.I. HARRIS ET AL.
rheumatoid arthritis receiving nonsteroidal anti-inflammatory drugs. A randomized, double-blind, placebo-controlled trial. Ann Intern Med. 1995;123:241-249.
parecoxib sodium: A potent and selective inhibitor of COX-2 for parenteral administration. J M e d Chem. 2000;43:1661-1663.
5. Goldstein JL, Silverstein FE, Agrawal NM, et al. Reduced risk of upper gastrointestinal ulcer complications with celecoxib, a novel COX-2 inhibitor. A m J Gastroenterol. 2000;95:1681-1690.
13. Kuss M, Mehlisch D, Bauman A, et al. Analgesic activity of single IV doses of parecoxib and ketorolac in postoperative dental pain. Poster presented at: 9th World Congress on Pain (WCP); August 22-27, 1999; Vienna, Austria; Board No. 249.
6. Lipsky PE, Abramson SB, Crofford L, et al. The classification of cyclooxygenase inhibitors. J Rheumatol. 1998;25(Suppl 12): 2298-2303.
14. Toradol® (ketorolac tromethamine). In: Physicians' Desk Reference ®. Montvale, NJ: Medical Economics Co; 2001:2789-2792.
7. Seibert K, Zhang Y, Leahy K, et al. Pharmacological and biochemical demonstration of the role of cyclooxygenase-2 in inflammation and pain. Proc Natl Acad Sci U S A. 1994;9l:12013-12017. 8. Vane JR. Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nat New Biol. 1971; 231:232-235. 9. Davies P, Bailey PJ, Goldenburg MM, Ford-Hutchinson AW. The role of arachidonic acid oxygenation products in pain and inflammation. Annu Rev lmmunol. 1984;2:335-357. 10. Garcia Rodriguez LA, Jick H. Risk of upper gastrointestinal bleeding and perforation associated with individual nonsteroidal anti-inflammatory drugs. Lancet. 1994;343:769-772. 11. Dubois RN, Abramson SB, Crofford L, et al. Cyclooxygenase in biology and disease. FASEB J. 1998;12:1063-1073. 12. Talley JJ, Bertenshaw SR, Brown DL, et al. N-[[(5-Methyl-3-phenylisoxazol-4-yl)phenyl]sulfonyl]propanamide, sodium salt,
15. Simon LS, Lanza FL, Lipsky PE, et al. Preliminary study of the safety and efficacy of SC-58635, a novel cyclooxygenase 2 inhibitor: Efficacy and safety in two placebo-controlled trials in osteoathritis and rheumatoid arthritis, and studies of gastrointestinal and platelet effects. Arthritis Rheum. 1998;41:1591-1602. 16. Traversa G, Walker AM, Ippolito FM, et al. Gastroduodenal toxicity of different nonsteroidal antiinflammatory drugs. Epidemiology. 1995;6:49-54. 17. Maliekal J, Elboim CM. Gastrointestinal complications associated with intramuscular ketorolac tromethamine therapy in the elderly. Ann Pharmacother. 1995;29: 698-701. 18. Wolfe PA, Polhamus CD, Kubik C, et al. Giant duodenal ulcers associated with the postoperative use of ketorolac: Report of three cases. Am J Gastroenterol. 1994;89: 1110-1111. 19. Estes LL, Fuhs DW, Heaton AH, et al. Gastric ulcer perforation associated with the use of injectable ketorolac. Ann Pharmacother. 1993;27:42-43.
1427
CLINICAL THERAPEUTICS®
20. Quigley EM, Donovan JP, Livingston WC. Ketorolac-related giant gastric ulcers. Am J Gastroenterol. 1994;89:631-632. 21. Lanza FL, Karlin DA, Yee JE Doubleblind placebo controlled endoscopic study comparing the mucosal injury seen with an orally and parenterally administered new nonsteroidal analgesic ketorolac tromethamine at therapeutic and supratherapeutic doses. Am J GastroenteroL 1987; 82:939a.
22. Nuutinen LS, Laitinen JO, Salomaki TE. A risk-benefit appraisal of injectable NSAIDs in the management of postoperative pain. Drug Saf 1993;9:380-393. 23. Lanza E Rack ME Lynn M, et al. An endoscopic comparison of the effects of etodolac, indomethacin, ibuprofen, naproxen, and placebo on the gastrointestinal mucosa. J Rheumatol. 1987;14" 338-341.
Address correspondence to: Jay L. Goldstein, MD, Professor of Medicine, University ot Illinois at Chicago, 840 South Wood Street (m/c 787), Room 734 CSB, Chicago, IL 60612. E-mail: [email protected]
1428