MARC members present

MARC secretariat present

Invited experts

Matters of administration

Matters arising

Pharmacovigilance issues

Review of the cardiovascular safety of the cyclooxygenase-2 selective inhibitors

Reference Material

Background Information

Introductory Comments

Discussion on individual COX-2 inhibitors

Celecoxib (Celebrex)
Etoricoxib (Arcoxia)
Valdecoxib (Bextra)
Parecoxib (Dynastat)
Lumiracoxib (Prexige)
Meloxicam (Mobic)
Rofecoxib (Vioxx)

General discussion on the COX-2 inhibitors

Questions posed by the Chair to the Committee
Discussion with a Consultant Rheumatologist
Peri-operative use of COX-2 inhibitors
Familial Adenomatous Polyposis (FAP) Intensive Medicines Monitoring Programme (IMMP)
Concurrent use of COX-2 inhibitors with Aspirin

Benefit-risk of all COX-2 inhibitors

Conclusions

General Conclusions

Recommendations

Celecoxib (Celebrex)
Etoricoxib (Arcoxia)
Valdecoxib (Bextra)
Parecoxib (Dynastat)
Lumiracoxib (Prexige)
Meloxicam (Mobic)
Rofecoxib (Vioxx)
General Recommendations

MARC members present

MARC secretariat present

Invited experts

1.

Matters of administration

1.1

Welcome and apologies

1.2

Minutes of the 120^th MARC meeting

1.3

Dates of 2005 MARC meetings

1.4

Conflicts of interest

2.

Matters arising

2.1

Report on actions arising from the 120^th MARC meeting

3.

Pharmacovigilance issues

3.1

Review of the cardiovascular safety of the cyclooxygenase-2 selective inhibitors

3.1.1

Reference Material

1. Minutes of the meeting between the MARC Chair and the Medsafe Pharmacovigilance team. 11 February 2005.
2. Medsafe Media Statement (22 February 2005). Ministry of Health Issues Strong Warning on COX-2 Inhibitors.
3. Medsafe "Dear Health Professional" fax (22 February 2005). Medsafe advises review of all patients on COX-2 inhibitors.
4. MIMS New Ethical (November 2004). Assessing cardiovascular risk and treatment benefit.
5. Ratified COX-2 Resolutions of the 238^th Meeting of the Australian Drug Evaluation Committee 3^rd to 4^th February 2005.
6. Jessamine S (2005). Notes from COX-2 Agenda item for Feb 2005 ADEC Meeting
7. Harrison-Woolrych M et al (2005). Incidence of thrombotic cardiovascular events in patients taking celecoxib compared with those taking rofecoxib: interim results from the New Zealand Intensive Medicines Monitoring Programme.

8. TGA Media Statement (10 February 2005). Regulator takes tough action on arthritis drugs.
9. McEwen J (2005). Expanded information on COX-2 inhibitors for doctors and pharmacists (amended).
10. EMEA Public Statement (17 February 2005). European Medicines Agency announces regulatory action on COX-2 inhibitors.
11. Duff G (February 2005). CSM "Dear Colleague" letter. Updated Advice on the Safety of Selective COX-2 inhibitors.
12. Health Canada Media Release (15 February 2005). Health Minister Ujjal Dosanjh gives a teleconference on drug safety and COX-2 drugs.

13. Aw T et al (2005). Meta-analysis of Cyclooxygenase-2 Inhibitors and Their Effects on Blood Pressure. Arch Intern Med 165:1-7.
14. Graham D et al (2005). Risk of acute myocardial infarction and sudden cardiac death in patients treated with cyclooxygenase-2 selective and non-selective non-steroidal anti-inflammatory drugs: nested case-control study. Lancet published online January 25 2005.
15. Kimmel S et al (2005). Patients Exposed to Rofecoxib and Celecoxib Have Different Odds of Nonfatal Myocardial Infarction. Annals of Internal Medicine 142:157-164.
16. Bresalier R et al (2005). Cardiovascular Events Associated with Rofecoxib in a Colorectal Adenoma Chemoprevention Trial. New England Journal of Medicine 352. Accessed online February 15 2005.
17. Solomon S et al (2005). Cardiovascular Risk Associated with Celecoxib in a Clinical Trial for Colorectal Adenoma Prevention. New England Journal of Medicine 352. Accessed online February 15 2005.
18. Nussmeier N (2005). Complications of the COX-2 Inhibitors Parecoxib and Valdecoxib After Cardiac Surgery. New England Journal of Medicine 352. Accessed online February 15 2005.
19. Levesque L et al (2005) The Risk for Myocardial Infarction with Cyclooxygenase-2 Inhibitors: A Population Study of Elderly Adults. Annals of Internal Medicine 142 (7): 481-489.
20. DTB Bulletin. Taking stock of coxibs. Vol 43, No. 1 January 2005
21. Maxwell S and Webb D (2005). COX-2 inhibitors - important lessons learned. Lancet published online 25 January 2005.
22. Finckh A and Aronson M (2005). Editorial. Cardiovascular Risk of Cyclooxygenase-2 Inhibitors: Where We Stand Now. Annals of Internal Medicine 142(3): 212-214.
23. Drazen J (2005). Editorial. COX-2 Inhibitors - A Lesson in Unexpected Problems. New England Journal of Medicine 352. Accessed online February 15 2005.
24. Psaty B and Furberg D (2005). Editorial. COX-2 Inhibitors - Lessons in Drug Safety. New England Journal of Medicine 352. Accessed online February 15 2005.
25. Topol E (2005). Editorial. Arthritis Medicines and Cardiovascular Events - "House of Coxibs". JAMA 293 (3): 366-368.
26. Laine Loren (2004). Proton Pump Inhibitor Co-Therapy With Nonsteroidal Anti-Inflammatory Drugs - Nice or Necessary? Reviews in Gastroenterological Disorders 4 (suppl 4): S33-S41.
27. Cryer B (2004). Editorial. COX-2 specific Inhibitor or Proton Pump Inhibitor Plus Traditional NSAID: Is Either Approach Sufficient for Patients at Highest Risk of NSAID-Induced Ulcers? Gastroenterology 127: 1256-1262
28. R Savage (2005). COX-2 Inhibitors - When Should They Be Used in the Elderly? Drugs Aging 2005; 22 (3).

29. Pengilley A (2005). Request for ADEC advice re: celecoxib. February 2005
30. Report for ADEC (2005). Clinical evaluation of cardiovascular safety of celecoxib.
31. MAAC pre-registration evaluation report of Celebrex
32. Pre-ADEC response from Pfizer re: celecoxib
33. New Zealand Celebrex data sheet

34. Pengilley A (2005). Request for ADEC advice re: etoricoxib. February 2005
35. Report for ADEC (2005). Clinical evaluation of cardiovascular safety of etoricoxib
36. MAAC pre-registration evaluation report of Arcoxia
37. Pre-ADEC response from Merck Sharp & Dohme re: etoricoxib
38. New Zealand Arcoxia data sheet

39. Pengilley A (2005). Request for ADEC advice re: valdecoxib. February 2005
40. Report for ADEC (2005). Clinical evaluation of cardiovascular safety of valdecoxib.
41. MAAC pre-registration evaluation report of Bextra
42. Pre-ADEC response from Pfizer re: valdecoxib
43. New Zealand Bextra data sheet

44. Pengilley A (2005). Request for ADEC advice re: parecoxib. February 2005
45. Report for ADEC (2005). Clinical evaluation of cardiovascular safety of parecoxib.
46. MAAC pre-registration evaluation report of Dynastat
47. Pre-ADEC response from Pfizer re: parecoxib
48. New Zealand Dynastat data sheet

49. Pengilley A (2005). Request for ADEC advice re: meloxicam. February 2005
50. Report for ADEC (2005). Clinical evaluation of cardiovascular safety of meloxicam.
51. MAAC pre-registration evaluation report of Mobic
52. Pre-ADEC response from Boehringer Ingleheim re: meloxicam
53. New Zealand Mobic data sheet

54. Pengilley A (2005). Request for ADEC advice re: lumiracoxib. February 2005
55. Report for ADEC (2005). Clinical evaluation of cardiovascular safety of lumiracoxib.
56. MAAC pre-registration evaluation report of Prexige
57. Pre-ADEC response from Novartis re: lumiracoxib

58. Submission from Merck Sharp and Dohme

• Vioxx FDA Advisory Committee Background Information.
• Arcoxia FDA Advisory Committee Background Information.
• Clinical Study Report EDGE Study MK-0663 (etoricoxib) Protocol 061 October 2004.
• European Union Summary of Product Characteristics for Arcoxia.
• European Medicines Agency (EMEA) Public Statement regarding COX-2 Inhibitors 17 February 2005.
• EMEA Questions and Answers on COX-2 Inhibitors 17 February 2005.
• European Union "Dear Doctor" letter.

59. Submission from Novartis.

• Lumiracoxib. Background Document for Novartis Presentation to FDA Advisory Committee (February 16-18, 2005).
• Joint Meeting of the Arthritis and the Drug Safety and Risk Management Advisory Committees. Gastrointestinal and Cardiovascular Safety of Lumiracoxib, Ibuprofen and Naproxen. February 17, 2005.
• Clinical Development and Medical Affairs. Lumiracoxib - Summary of Cardiovascular Safety.
• Proposed Lumiracoxib Post-Approval Enhanced Pharmacovigilance Plan.

60. Submission from Pfizer

• Table of Contents and Executive Summary
• Discussion of Medsafe Proposal and Clinical Data.
• Benefit-Risk Analysis.
• Risk Management Plan.
• Proposed Labelling.
• FDA Briefing Document.
• Celecoxib EMEA Response.
• Valdecoxib EMEA Response.
• Parecoxib EMEA Response.
• Long Term Trials Summary for Celecoxib & NEJM APC Publication
• Medical Index Data IMS
• Quantitative Assessment of the Cardiovascular and Gastrointestinal Risk/Benefit Profile of Celecoxib Compared to Individual NSAIDs.

61. Submission from Brookfields Lawyers on behalf of Boehringer Ingelheim

• Summary of Boehringer Ingelheim's position.
• Supporting Medical Argument.

Correspondence received by Medsafe

62. Correspondence from Professional Organisations

• Submission from The New Zealand Rheumatology Association.

63. Correspondence from Health Professionals

• Medsafe responses to health professional correspondence.

64. Correspondence from consumers

• Medsafe responses to consumer correspondence

65. Dai C et al (2005). National Trends in Cyclooxygenase-2 Inhibitor use Since Market Release - Nonselective Diffusion of a Selectively Cost-Effective Innovation. Arch Intern Med 2005; 165:171-177.
66. Battistella M et al (2005). Risk of Upper Gastrointestinal Haemorrhage in Warfarin Users Treated With Nonselective NSAIDs or COX-2 Inhibitors. Arch Intern Med. 2005;165:189-192.
67. Van Hecken A et al (2000). Comparative Inhibitory Activity of Rofecoxib, Meloxicam, Diclofenac, Ibuprofen, and Naproxen on COX-2 versus COX-1 in Healthy Volunteers. J Clin Pharmacol. 2000; 40:1109-1120.
68. Pfizer Inc. Protocol IQ5-97-02-001 Re: Celebrex/Celecoxib.
69. Walls J. on behalf of Pfizer. Email. Celebrex, Bextra and Dynastat - Proposed Data Sheets.
70. Pfizer Inc. Dynastat New Zealand Data Sheet - Draft of Changes Proposed by Pfizer.
71. Pfizer Inc. Bextra New Zealand Data Sheet - Draft of Changes Proposed by Pfizer.
72. Pfizer Inc. Celebrex New Zealand Data Sheet - Draft of Changes Proposed by Pfizer.
73. Submission from the New Zealand Society of Anaesthetists Inc re: The Distribution of COX-2 Inhibitor Drugs.
74. Letter from a consumer.
75. Letter from a health professional.
76. Letter from a consumer to the Minister of Health. 
77. Letter from a consumer to the Minister of Health.
78. Letter from a health professional.
79. Fax from a health professional.
80. Letter from a consumer.
81. Letter from a health professional.
82. Letter from a health professional.
83. Fax from a health professional.
84. Open Letter to New Zealand Medsafe from the New Zealand Pain Society re: A possible withdrawal of all COX-2 selective non-steroidal anti-inflammatory drugs from the New Zealand market. March 9, 2005.
85. Letter from a health professional.
86. Boehringer Ingelheim. Presentation to Medsafe March 11, 2005. Pharmacological and clinical study data as a basis for assessment of the comparative cardiovascular toxicity profile of meloxicam.

87. Jüni et al (2004). Risk of cardiovascular events and rofecoxib: cumulative meta-analysis. Lancet published online November 5, 2004
88. Ray W et al (2002). COX-2 selective non-steroidal anti-inflammatory drugs and the risk of serious coronary heart disease. Lancet 360:1071-1073.
89. White et al (2002). Comparison of thromboembolic events in patients treated with celecoxib, a cyclooxygenase-2 specific inhibitor, versus ibuprofen or diclofenac. Am J Cardiol. 2002 Feb 15; 89(4): 425-30.
90. Farkouh M et al (2004). Comparison of lumiracoxib with naproxen and ibuprofen in the Therapeutic Arthritis Research and Gastrointestinal event Trial (TARGET), cardiovascular outcomes: randomised controlled trial. Lancet 364:675-684
91. White W et al (2004). Effects of the Cyclooxygenase-2 specific inhibitor Valdecoxib versus nonsteroidal anti-inflammatory agents and placebo on cardiovascular thrombotic events in patients with arthritis. American Journal of Therapeutics 11.244-250.
92. Bombardier C et al for the VIGOR study group (2000). Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. The New England Journal of Medicine; 342:1520-1528.

3.1.2

Background Information

• COX-2 agents are not recommended for routine use in patients with rheumatoid arthritis (RA) or osteoarthritis (OA) except in circumstances where the patient is at "high risk " of developing a serious gastrointestinal adverse effect from other standard non-steroidal anti-inflammatory agents;
• COX-2 agents should not be routinely prescribed to patients at high risk of cardiovascular events (absolute risk of event >15-20% over 5 years) as there remains uncertainty about the safety of these agents when used in this group of patients;
• Prescribing COX-2 agents to patients already taking aspirin cannot be justified on current evidence;
• The increased benefit associated with use of COX-2 agents in patients at high risk of serious gastrointestinal adverse effects (estimated as between a 4 and 8 times reduction in serious events), may outweigh the increased risk of cardiovascular events in patients at low risk of myocardial infarction.

Therefore, with the agreement of MARC members, a meeting was convened between the Medsafe pharmacovigilance team and the MARC Chair on 11 February 2005. Prior to this meeting, Medsafe, the MARC Chair and all members of the MARC were provided with TGA-commissioned external evaluations of the data supplied by product sponsors on each of the COX-2 inhibitors, as well as recent published literature, the recommendations of ADEC and the company responses to the TGA commissioned evaluations.

• Medsafe and the MARC Chair accept that although there is clear evidence that an increased cardiovascular risk is a class effect, the level of cardiovascular risk associated with each individual COX-2 inhibitor cannot be defined from the available evidence. However, the available evidence implicates several COX-2 inhibitors as having an increased cardiovascular risk associated with short-term and long-term use, and in patients at high and low risk of cardiovascular disease. Therefore, Medsafe and the MARC Chair consider that it is not possible to identify any particular patient group, or indication for use, as being free from an increased cardiovascular risk associated with the use of a COX-2 inhibitor.
• Similarly, it is difficult to clearly define which subgroup of patients is likely to benefit from the gastrointestinal sparing effect of COX-2 inhibitor therapy, and in whom the benefits of treatment may outweigh the risks.
• After evaluation of the available evidence, it is concluded that the potential risks outweigh the potential benefits of COX-2 inhibitor therapy for most patients. In view of the inability of current evidence to quantify the risk associated with each individual product or to clearly establish 'at risk' populations, a cautious approach to the use of these medicines is warranted.

1. The MARC Chair will contact all MARC members to inform them of the preliminary conclusions reached by Medsafe and the MARC Chair. Members will be informed that the cardiovascular safety of the COX-2 inhibitors will be placed on the agenda for discussion at the next MARC meeting scheduled for Tuesday 15 March 2005.
2. Whilst awaiting final review of the evidence by the full MARC, the following precautionary advice should be communicated to prescribers:
   • COX-2 inhibitor treatment should be stopped immediately in all patients with a previous history of MI or stroke or at high cardiovascular risk (e.g. positive family history, diabetes, smoking, hypercholesterolaemia or hypertension).
   • All other patients should discuss alternative treatment options with their GP at their next scheduled appointment.
   • Medsafe recommends that prescribers refer to the recently issued Best Practice Advocacy Centre (BPAC) guidance for advice on the use of alternative analgesia (NSAIDs-strategies for minimising harm - BPAC 2004). This publication also provides advice on the management of patients at high risk of gastrointestinal ulceration or bleeding.
3. Medsafe should inform the COX-2 inhibitor product sponsors of the outcome of the review undertaken by the Medsafe and the MARC Chair. Medsafe should inform the product sponsors of the potential options available under the Medicines Act (1981) to manage the risk associated with the COX-2 inhibitors. Product sponsors should be given the opportunity to respond to the concerns identified by Medsafe and the MARC Chair and should be requested to provide any additional safety data to Medsafe by 1 March 2005 for consideration by the MARC on 15 March 2005.
4. Medsafe should review the results of regulatory reviews currently being undertaken by the FDA and the EMEA and provide this information to the MARC for consideration on 15 March 2005.
5. It is anticipated that the MARC will issue definitive advice on the cardiovascular safety of the COX-2 inhibitors following their March meeting. Medsafe should be prepared to communicate this advice to prescribers and consumers via a media statement and a "Dear Health Professional" letter.

1. The New Zealand Rheumatology Association made the following recommendations ⁶²:

• "COX-2 inhibitors should continue to be available for patients in whom the increase in absolute risk of thrombotic cardiovascular events is low AND for whom there has been a substantial benefit from COX-2 inhibitors AND this benefit was not derived from use of traditional NSAIDs or other treatments, or these other alternatives were not tolerated."
• "It should be the responsibility of the prescribing doctor to make a risk-benefit assessment in each individual case, as already occurs with other drugs such as prednisone and third- generation oral contraceptives."

2. The New Zealand Society of Anaesthetists concluded⁷³:

• "That the decision regarding the permanent revocation of the consent to distribute these drugs in New Zealand be modified to allow their continued use in the peri-operative period.
• That direct-to-consumer advertising of this class of drug be halted.
• That patients be empowered to make an informed choice about using these drugs in the peri-operative period.
• That these decisions should be reviewed when and if further data becomes available."

3. The New Zealand Pain Society wrote that⁸⁴:

• "A total withdrawal of COX-2 selective NSAIDs based on adverse events from studies of their long term use in patients at potential risk of such events, will have a significant impact on a whole group of patients in which a short-term course with a coxib seems highly justified based on our available evidence.
• A continuing attention to and documentation of any untoward effects of this class of drugs and an ongoing educational process of the medical community is deemed prudent. With an increasing database of the efficacy and cost-efficiency of coxibs in short-term analgesia, an ongoing revision of the justification of their use should take place.
• We strongly advocate that Medsafe leave coxibs available in New Zealand for continued short- term use in select patient groups (e.g. in peri-operative and in terminal cancer patients)."

4. Correspondence from consumers and health professionals generally supported the continued availability of COX-2 inhibitors^{63, 64}. It was acknowledged that individual risk-benefit assessments were needed to ensure safe use of these medicines. A number of consumers expressed the opinion that they would be prepared to accept an increased cardiovascular risk in return for maintenance of adequate analgesia

• Seeking further information or undertaking further consultation with interested parties.
• Requiring data sheet changes that might include the addition of new contraindications, warnings and precautions and/or removal of indications.
• Producing a "Dear Health Professional" letter to communicate with prescribers.
• Writing an article for publication in Prescriber Update.
• Regulatory action under the Medicines Act 1981 such as:

• Issuing a Section 36 notice to sponsors requesting further safety information.
• Revocation of product consents and re-registration of the product under Section 23 with new prescribing restrictions and/or monitoring requirements.
• Revocation or suspension of product license under Section 35.

3.1.3

Introductory Comments

Members agreed that it is difficult to assess what level of cardiovascular risk is acceptable, particularly as COX-2 inhibitors are not disease-modifying medicines.

3.1.4

Discussion on individual COX-2 inhibitors

3.1.4.1

Celecoxib (Celebrex)

Celecoxib was approved for use in New Zealand in 1998. In New Zealand, celecoxib is indicated for:

1. Symptomatic treatment of pain and inflammation in osteoarthritis and rheumatoid arthritis.
2. For the management of acute pain and treatment of primary dysmenorrhoea in adults.
3. To reduce the number of adenomatous polyps in Familial Adenomatous Polyposis (FAP) as an adjunct to usual care.

The approved dosage regimens are:

1. Osteoarthritis: Recommended dose is 200mg. A dose of 200mg bd can be used if required.
2. Rheumatoid arthritis: 200-400mg per day in two divided doses.
3. FAP: 400mg bd
4. Acute pain and dysmenorrhoea: 400mg as a single dose on the first day followed by 200mg once daily on subsequent days. An additional dose of 200mg can be given on any one day if required. Doses up to 600mg daily have been studied for these indications. Celebrex can be administered up to 2 hours prior to surgery.

Pfizer estimates that approximately 11,600 patients received Celebrex in New Zealand between January 2004 and December 2004. The average daily dose is estimated to be 222mg per day.

What is the evidence for cardiovascular adverse effects attributable to COX-2 inhibition for celecoxib?

• The members noted that the APC trial¹⁷ provides robust evidence that, at high doses, celecoxib is associated with an increased risk of myocardial infarction, stroke and cardiac death compared to placebo. In the APC trial, the RR associated with celecoxib 400mg bd is 3.4 (CI 95% 1.4-8.3), and the RR associated with celecoxib 200mg bd is 2.5 (CI 95% 1.0-6.3)
• However, the Committee noted that there is an extensive amount of published epidemiological data that has concluded that celecoxib is not associated with an increased risk of cardiovascular disease compared to other Non-Steroidal Anti-inflammatory Drugs (NSAIDs). Members discussed the following studies:
  • Long-term randomised controlled trials, including PreSAP⁶⁰, ADAPT⁶⁰ and CLASS⁶⁰ did not show any increased cardiovascular risk associated with celecoxib.
  • The White et al (2002)⁹¹ re-analysis of the CLASS data did not identify an increased cardiovascular risk. However, it is noted that assessment of cardiovascular adverse events was only a secondary endpoint of the study and the study was not adequately powered to address this risk. It was also noted that most of the data available from the CLASS study was of less than 6 months duration. It was noted that the CLASS study showed no difference between the aspirin and the non-aspirin groups.
  • In the CAESAR study⁶⁰ 458 patients were randomised to receive celecoxib or diclofenac and were followed up for a maximum of 1 year. There was no increase in cardiovascular adverse events in the celecoxib arm, but there were more withdrawals in the celecoxib group. Cardiovascular endpoints were not pre-specified or adjudicated in this study.
  • In the SUCCESS 1 study⁶⁰ 13,000 patients were randomised to receive celecoxib, naproxen or diclofenac for 12 weeks. There was no difference in event rates; however, the number of events was small. It is noted that cardiovascular endpoints were not pre-specified or adjudicated in this study. This study was not accepted for publication.
  • Levesque L et al¹⁹ conducted a nested case-control study using the Quebec Administrative Health database. This study involved nearly 114,000 elderly people who had received a prescription for any NSAID including a COX-2 inhibitor. Members noted that although there is no evidence of an increased cardiovascular risk with celecoxib compared to other NSAIDs (apart from rofecoxib), there is the potential for confounding with respect to use of aspirin or other over-the-counter (OTC) NSAIDs.

• The members noted that APC¹⁷ is the only study that is of sufficient duration to assess an effect of duration of treatment on the level of cardiovascular risk. In the APC study an increase in cardiovascular risk became evident only after approximately 12 months of use. However, it was noted that these results are still preliminary at this point, as the monitoring is not complete. Therefore, care must be taken in the interpretation of this study.
• The members noted that the APC study also provides some evidence that cardiovascular risk is increased with increasing doses of celecoxib. However, more data are needed to more clearly elucidate the clinical significance of the dose relationship.
• Based on the data reviewed, the members agreed that any treatment with celecoxib should be for the shortest time possible at the lowest dose possible. However, it was noted that some patients might require higher doses for a short period of time (e.g. in migraine or sports injuries); and other patients may require chronic treatment (such as rheumatoid arthritis or FAP). A careful risk assessment would be required for each individual patient before prescribing celecoxib.

3.1.4.2

Etoricoxib (Arcoxia)

Etoricoxib was approved for use in New Zealand in 2002. Etoricoxib is indicated for use in:

1. Acute and chronic treatment of the signs and symptoms of osteoarthritis and rheumatoid arthritis
2. Treatment of acute gouty arthritis
3. Relief of acute pain
4. Relief of chronic musculoskeletal pain

The approved dosage regimens are:

Osteoarthritis: 60mg once daily
Rheumatoid arthritis: 90mg once daily
Acute gouty arthritis: 120mg once daily - for the acute symptomatic period only.
Acute pain: 120mg once daily for the acute symptomatic period only.
Chronic musculoskeletal pain: 60mg once daily

MSD estimates that in the year January 2004-December 2004 a monthly average of 1400 units of 30 x 60mg, 645 units of 30 x 90mg and 1750 units of 10 x 120mg of etoricoxib were distributed in New Zealand.

• Members commented that the available data with respect to etoricoxib use and the occurrence of thrombotic cardiovascular adverse events are very limited. MSD's submission⁵⁸ notes that placebo controlled data are limited because of the need for active controls in arthritis trials.
• In an analysis of all chronic exposure studies, there was a statistically significant increase in Antiplatelet Trialists' Collaboration (APTC) endpoints (RR 2.72, CI 95% 1.18 - 6.27) in the etoricoxib group compared to naproxen. However, it was acknowledged that number of events in each group was small.
• Members noted that in the EDGE study⁵⁸ there was no evidence of an increase in cardiovascular risk compared to diclofenac. However, although EDGE was designed to be a 12-month trial, only 31% of patients remained in the etoricoxib arm at the end of 12 months. Hence the data on the chronic use of etoricoxib are very limited.
• Members commented that the EDGE study did not provide convincing evidence of etoricoxib's superiority over traditional NSAIDs with respect to gastrointestinal safety. It was noted that there was no difference in the incidence of complicated ulcers in patients taking aspirin between the diclofenac and etoricoxib groups.
• Members noted that in the EDGE trial the incidence of hypertension-related adverse events was 11.7% in the etoricoxib group vs. 5.9% in the diclofenac group.
• Members considered that the available data provide evidence of a signal for increased cardiovascular risk associated with etoricoxib. However, there are insufficient data to enable quantification of this increase in risk.
• The members recommended that etoricoxib be contraindicated in hypertension that is not well controlled. When prescribed in controlled hypertension close monitoring is essential.

3.1.4.3

Valdecoxib (Bextra)

Valdecoxib was approved for use in New Zealand in 2003. Valdecoxib is indicated for use in:

1. Symptomatic treatment of osteoarthritis and rheumatoid arthritis
2. Treatment of acute pain, including that related to primary dysmenorrhoea and minor dental procedures
3. Treatment of mild to moderate post-operative pain.

The approved dosage regimens are:

Symptomatic treatment of osteoarthritis and rheumatoid arthritis: 10-20mg once daily
Treatment of acute pain including that related to primary dysmenorrhoea and minor dental procedures: 40mg once daily
Mild to moderate post-operative pain. 40mg once daily. On the first day of treatment an additional 40mg may be taken

Pfizer estimates that, in New Zealand, approximately 7000 prescriptions for Bextra were written between January 2004 and December 2004 of which 70% were for 20mg per day and 30% were for 40mg per day.

• Members discussed anecdotal reports that valdecoxib is being used extensively in the peri-operative setting.
• Members noted that the submissions from the NZ Society of Anaesthetists and some health professionals indicate that valdecoxib use might provide advantages over alternative agents for specific surgical situations. It was suggested that this might include use in patients undergoing tonsillectomy, gynaecological and breast surgery as well as orthopaedic, neurological and hepatic surgery. It was noted that use of COX-2 inhibitors, rather than opiates, in day surgery might facilitate earlier recovery and discharge.
• However, members expressed concern that patients need to be fully informed of the risks associated with use of medicines in the peri-operative situation. Members were concerned that this may not be happening unless these discussions are undertaken during the pre-operative assessment of risk. It was noted that it is the responsibility of the individual physician to make an assessment of risk:benefit.
• There are two published studies^{18, 91} in CABG patients taking valdecoxib and parecoxib that have detected an increase in cardiovascular risk compared to placebo. In the recently published Nussmeir et al¹⁸ study there is evidence that this difference in risk occurs from the first day of treatment. It was noted that patients undergoing CABG are at very high risk of cardiovascular events.
• The members discussed the sponsor's view that the increased risk seen in the CABG studies is an aberration, seen only in very high-risk patients, and that these results do not apply to the general population of surgical patients.
• It was noted that, although the sponsor supplied a large amount of data on the use of valdecoxib (and parecoxib) in surgical populations other than CABG, these studies were underpowered to detect a doubling of cardiovascular risk. Therefore, the Committee considered these studies to be uninformative with respect to elucidating the level of cardiovascular risk associated with valdecoxib use.
• It was noted that valdecoxib is also indicated for use in the treatment of osteoarthritis, rheumatoid arthritis and acute pain, but there are very limited data available on the long-term use of valdecoxib. There are also insufficient data regarding subjects with multiple cardiac risk factors, as well as subjects who are at low risk for cardiac adverse events. There are also limited data regarding the use of COX-2 inhibitors in surgical procedures other than cardio-pulmonary.
• The Committee concluded that there is clear evidence of an increased risk of cardiovascular events in patients at high cardiovascular risk undergoing high-risk surgical procedures. However, there are not enough data available to assess the risk associated with use of valdecoxib in other patient groups and in other surgical procedures.
• Following considerable discussion the consensus among the Committee members was that, given the current limitations of the evidence, a precautionary stance is preferable.

3.1.4.4

Parecoxib (Dynastat)

Parecoxib was approved for use in New Zealand in 2003. Parecoxib is indicated for use as:

A single peri-operative dose for the management of post-operative pain at a dose of 40mg IV or IM.

Pfizer estimates that, in New Zealand, 9,665 patients received parecoxib between January 2004 and December 2004.

• Members noted that the submissions from the NZ Society of Anaesthetists and some health professionals indicate that parecoxib use might provide advantages over alternative agents for specific surgical situations. It was suggested that this might include use in patients undergoing tonsillectomy, gynaecological and breast surgery as well as orthopaedic, neurological and hepatic surgery. It was noted that use of COX-2 inhibitors rather than opiates in day surgery, may facilitate earlier recovery and discharge.
• However, members expressed concern that patients need to be fully informed of the risks associated with use of medicines in the peri-operative situation. Members were concerned that this may not be happening unless these discussions are undertaken during the pre-operative assessment of risk. It was noted that it is the responsibility of the individual physician to make an assessment of risk:benefit.
• There are two published studies^{18, 91} in CABG patients taking valdecoxib and parecoxib that have detected an increase in cardiovascular risk compared to placebo. In the recently published Nussmeir et al¹⁸ study there is evidence that this difference in risk occurs from the first day of treatment. It was noted that patients undergoing CABG are at very high risk of cardiovascular events.
• The members discussed the sponsor's view that the increased risk seen in the CABG studies is an aberration, seen only in very high-risk patients, and that these results do not apply to the general population of surgical patients.
• It was noted that, although the sponsor supplied a large amount of data on the use of parecoxib (and valdecoxib) use in surgical populations other than CABG, these studies were underpowered to detect a doubling of cardiovascular risk. Therefore the Committee considered these studies to be uninformative with respect to elucidating the level of cardiovascular risk associated with parecoxib use.
• The Committee concluded that there is clear evidence of an increase risk of cardiovascular events in patients at high cardiovascular risk undergoing high-risk surgical procedures. However, there are not enough data available to assess the risk associated with use of parecoxib in other patient groups and in other surgical procedures.
• Following considerable discussion the consensus among the Committee members was that, given the current limitations of the evidence, a precautionary stance is preferable.

3.1.4.5

Lumiracoxib (Prexige)

Lumiracoxib was approved for use in New Zealand in 2004. It is approved for use in:

1. Symptomatic treatment of osteoarthritis.
2. Treatment of acute pain
3. Treatment of primary dysmenorrhoea

Lumiracoxib is not yet marketed in New Zealand.

• The members noted from in-vitro studies that lumiracoxib is the most COX-2 selective of the agents being considered, which some members considered might be reflected in the relatively high degree of gastrointestinal safety of lumiracoxib compared to naproxen, as demonstrated in the TARGET study^90.
• It was noted that the TARGET study⁹⁰ consisted of two separate sub-studies - lumiracoxib versus naproxen and lumiracoxib versus ibuprofen. There were significant differences in the characteristics of the subjects in these two cohorts, with respect to race, gender, baseline cardiovascular risk, aspirin use and use of other cardiovascular medications. Thus, members considered that it was inappropriate to combine the results of the two sub-studies.
• The members considered that the data do not justify the statement made in the sponsor's submission that "an increase in COX-2 selectivity does not automatically lead to an increase in cardiovascular risk."
• The members commented that, although the sponsor states that "the TARGET study includes high numbers of high cardiovascular risk patients" ⁵⁹, the TARGET study had strict exclusion criteria that excluded patients at high risk of cardiovascular disease who were not on aspirin; patients with a history of myocardial infarction; stroke or new onset angina within 6 months of starting the study; patients with a history of stroke who were not on aspirin; and patients with ECG evidence of silent myocardial ischaemia.
• In the ibuprofen arm of the TARGET study, there was no statistically significant increase in cardiovascular risk.
• In the naproxen sub-study, in patients not taking aspirin, there was a non-statistically significant increase in the risk of myocardial infarction in the lumiracoxib group versus the naproxen group (HR 2.37, CI 95% 0.74-7.55). As the TARGET study excluded patients at high risk of cardiovascular adverse events who were not taking aspirin, the non-aspirin group is a low-risk population.
• Despite the fact that the TARGET study had pre-specified cardiovascular endpoints, and enrolled more than 18,000 patients, it was still underpowered to detect an increase in cardiovascular risk. This is evident from the small number of cardiovascular events in each of the study arms.
• The members noted that the TARGET study provides good evidence that lumiracoxib is associated with fewer gastrointestinal adverse events compared to non-selective NSAIDs. However, this gastrointestinal benefit is negated when lumiracoxib is combined with low-dose aspirin.
• The members noted that lumiracoxib is significantly worse than NSAID comparators for renal and hepatic endpoints.
• Therefore, after reviewing the available evidence, the MARC considered that there is a signal for increased cardiovascular risk associated with lumiracoxib compared to naproxen.
• In summary, on the basis of the existing data, the Committee considered that there is inadequate evidence to support the claim for the cardiovascular safety of lumiracoxib.

3.1.4.6

Meloxicam (Mobic)

Meloxicam was approved for use in New Zealand in 1998. Meloxicam is approved for use in:

1. Symptomatic treatment of painful osteoarthritis
2. Symptomatic treatment of rheumatoid arthritis

The approved dosage regimens are:

Osteoarthritis and rheumatoid arthritis: 7.5mg per day

Boehringer Ingleheim estimates that, in New Zealand, approximately 8,770 patients received meloxicam in the year January 2004-December 2004.

• The members discussed the sponsor's submission that in-vitro meloxicam is less COX-2 selective than several of the other members of the COX-2 class of medicines and that structurally it is an oxicam rather than a coxib. However, the Committee noted that the sponsor had previously marketed meloxicam as a COX-2 inhibitor, and that there is considerable uncertainty regarding the clinical significance of in-vitro comparative measures of COX-1 and -2 inhibition.
• The members noted that, while long-term clinical trial safety data are not available for meloxicam, several Prescription Event Monitoring⁶¹ studies have been conducted and have not demonstrated any increased risk of cardiovascular adverse effects compared to celecoxib or rofecoxib. Overall, the body of evidence for meloxicam is more comparable to that of etoricoxib and valdecoxib than the other COX-2 inhibitors, in that the data are largely limited to short-term studies (less than 6 months). In the absence of long-term clinical trial data it is impossible to assess whether meloxicam use is associated with increased cardiovascular risk.
• The members considered that any restrictions placed on meloxicam should be consistent with the overall conclusion that there is a class effect for products exhibiting a predominant COX-2 inhibitory effect, and there are relative limitations of the data overall. Therefore, members agreed that meloxicam is to be considered along with the other COX-2 selective inhibitors for the purposes of assessing its cardiovascular risk.
• The Committee noted that preliminary data from an unpublished study by Graham et al was presented by the sponsor in their submission, but considered that it should not be included in the assessment of meloxicam.
• The members considered that, on the basis of the existing data, there is inadequate evidence in support of the claim of cardiovascular safety of meloxicam.

3.1.4.7

Rofecoxib (Vioxx)

Rofecoxib was approved for use in New Zealand in 2000. Rofecoxib was voluntarily withdrawn worldwide on 1 October 2004 due to concerns regarding its cardiovascular safety. Prior to its withdrawal, rofecoxib was approved for use in the following conditions:

• Acute and chronic treatment of the signs and symptoms of osteoarthritis.
• Acute and chronic treatment of the signs and symptoms of rheumatoid arthritis.
• Acute and chronic treatment of the signs and symptoms of juvenile rheumatoid arthritis in patients 2 years and older.
• Relief of pain.
• Treatment of primary dysmenorrhoea.
• Acute treatment of migraine attacks with or without aura

• Osteoarthritis: The recommended starting dose is 12.5 mg once daily. Some patients may receive additional benefit by increasing the dose to 25 mg once daily. The maximum recommended daily dose is 25 mg.
• Rheumatoid Arthritis: The recommended starting dose is 25 mg once daily. The maximum recommended daily dose is 25 mg.
• Juvenile Rheumatoid Arthritis: The recommended dose for paediatric patients 2 to 11 years of age is 0.6 mg/kg to a maximum of 25 mg once daily. To improve dosing accuracy for children weighing less than 40kg, the use of 12.5 mg/5 ml oral suspension (2.5 mg/ml) is recommended. The recommended dose for adolescent patients 12 to 17 years is 25 mg once daily. The maximum recommended daily dose is 25 mg.
• For the Relief of Acute Pain and Treatment of Primary Dysmenorrhoea: The recommended initial dose is 50mg once daily. Subsequent doses should be 25 to 50 mg once daily. The maximum recommended daily dose is 50 mg. Chronic use of VIOXX 50mg daily is not recommended (See Warnings and Precautions).
• Migraine attacks with or without aura: The recommended starting dose is 25mg once daily. Some patients may receive additional benefit by increasing the dose to 50mg once daily. The maximum recommended daily dose is 50mg. Chronic use of VIOXX 50mg daily is not recommended. (See Warnings and Precautions).

• The members discussed a number of studies including the VIGOR study⁹², the APPROVe study¹⁶, Ray W et al⁸⁸, Jüni et al⁸⁷ and Graham et al¹⁴. Members commented that there is an extensive body of evidence implicating rofecoxib in causing an increase in cardiovascular adverse events. There is also evidence that the risk is dose dependent.
• In the VIGOR study, there is evidence that the increased cardiovascular risk in the rofecoxib arm of the study was evident after approximately 1 month. In the APPROVe study, a difference in hypertension and heart failure was evident after about 5 months
• Members agreed that there is a need to be careful when assessing post-marketing meta-analyses from sponsors, where cardiovascular events are not the primary endpoints of the individual studies included in the meta-analyses.
• Members noted that the VIGOR study provided evidence for an early effect of rofecoxib whereas the APPROVe study did not show a difference with placebo until after 18 months of use. Members discussed the possibility that the APPROVe study may have been underpowered to show an early increase in risk. It was also noted that this difference might be due to a chance finding.
• Members concluded that no further discussion would be undertaken on rofecoxib unless the product sponsor informed Medsafe of its intent to remarket rofecoxib in New Zealand.

3.1.5

General discussion on the COX-2 inhibitors

3.1.5.1

Questions posed by the Chair to the Committee

In order to facilitate the discussion, the MARC Chair posed the following questions to the Committee:

1. Do you think that the experimental research suggests a potential pro-thrombotic effect for COX-2 inhibitors, in-vitro and in-vivo?

   The members noted that there is good evidence to justify COX-2 selectivity and the plausibility of a biological mechanism. The proposed hypothesis involves the balance between the inhibition of prostaglandin synthesis (PGI₂ mediates vasodilation and limits the initiation and progression of atherosclerosis) and thromboxane A₂ (TXA₂ stimulates platelet aggregation, vasoconstriction, endothelial cell migration, angiogenesis, and smooth muscle cell proliferation).

   The members noted that, although the results of VIGOR, APPROVe and APC can be explained on the basis of the hypothesis above, there is no evidence of a direct clinical correlation between in-vivo levels of prostacyclin and outcome at the present time. Evidence from VIGOR, APC and APPROVe, however, suggests that COX-2 inhibitors have a pro-thrombotic effect in-vivo.

   The members note that the proposed biological mechanism may not be the only mechanism of action for the increased cardiovascular risk associated with COX-2 inhibition. There is a complex cascade of effects involving multiple body systems.

   In summary, the Committee considered that there is a plausible biological mechanism to explain the pro-thrombotic effects of COX-2 inhibitors, and this is supported by evidence from some clinical studies. However, the evidence is not currently sufficient to make a definitive assessment.

2. Can the COX-2 inhibitors, and COX-2 selectivity, be ranked in a clinical sense?

   The members noted that there is evidence for a cascade of COX-2 selectivity in-vitro; however, the translation of the in-vitro cascade to a clinical setting is complex and individually variable.

   The members noted that there is no evidence regarding the degree of COX-2 selectivity in-vivo. Consequently, there is no evidence that the level of COX-2 selectivity correlates with analgesic efficacy or cardiovascular adverse effects in-vivo. The ratio and expression of COX-1 and COX-2 receptors varies in different tissues.

   The members noted that the in-vivo application of the COX-2 selectivity cascade is affected by individual variation in pharmacodynamics and pharmacokinetics, so that the in-vitro information has no clinical correlate on an individual risk basis.

   In contrast, the members noted that there is evidence that the COX-2 inhibitors can be ranked according to the degree of gastro-intestinal benefit conferred, and that this appears to correlate to some extent with the level of COX-2 selectivity. However, this is difficult to assess for the COX-2 inhibitors with the available data.

3. What is the evidence for a class effect of cardiovascular risk with COX-2 inhibition?

   Members considered that if two medicines in a class of medicines have been reliably shown to cause an increase in cardiovascular risk, and if the underlying hypothesis is related to the same mechanism of effect, then it is reasonable to assume a class effect for the increased risk, until proven otherwise.

   The Committee considered that the COX-2 inhibitors meet the conditions above, and that therefore there is an overall class effect for cardiovascular risk.

   The Committee considered that the onus is on the sponsors of the individual COX-2 inhibitors to disprove this conclusion.

3.1.5.2

Discussion with the Consultant Rheumatologist

A Consultant Rheumatologist with a background in research joined the meeting as an invited expert. 

The absolute cardiovascular risk with COX-2 inhibitors is small, but difficult to quantify with regards to 'at-risk' populations. Rheumatologists accept that the background cardiovascular risk is higher for rheumatoid arthritis patients compared to osteoarthritis patients, due to the inflammatory condition predisposing the patients to cardiovascular adverse events.

There is considerable variation between patients with regards to their response to individual agents from within the non-selective NSAID and COX-2 inhibitor classes. The reason for this variation is not adequately explained by clinical trials, but is clearly seen in clinical practice.

Treatment options for rheumatoid arthritis patients in New Zealand are limited. Many therapies are not subsidised, including the COX-2 inhibitors. Therefore, the patient population is self-selecting to some degree. Anecdotal evidence is that usage of COX-2 inhibitors is already decreasing.

Direct-to-consumer advertising (DTCA) has resulted in many more patients taking COX-2 inhibitors, some of who may not have tried other anti-inflammatory agents.

Consultant Rheumatologist rarely uses doses of greater than 200mg/day of celecoxib (or greater than 25mg/day of rofecoxib when it was available), and indicated that his colleagues generally use similar doses.

1. What is the benefit for rheumatology patients regarding the use of COX-2 inhibitors? There are no clear data available, however, there is a consensus among Rheumatologists in New Zealand as follows:

• Rheumatology patients take COX-2 inhibitors primarily if they cannot tolerate non-specific NSAIDs, or if they do not receive an adequate response with non-specific NSAIDs. For this group of patients it is not possible to return to treatment with non-specific NSAIDs alone.
• There is a group of patients for whom their quality of life, level of function and employment prospects would suffer should COX-2 inhibitors be removed from the market.
• For a group of patients the risk:benefit equation favours continuation of treatment with a COX-2 inhibitor.

2.  If the COX-2 inhibitors are not withdrawn from the market, how could their use be best restricted?

• A Special Authority would not be useful, because the majority of arthritis patients are managed primarily by their General Practitioner. As there are only 40 Rheumatologists in New Zealand a Special Authority is not practical.
• Education of prescribers and patients is important.
• Elimination of DTCA is important.

3. What is your view on the possible re-introduction of rofecoxib?

• Would like to see rofecoxib re-introduced.
• Etoricoxib and valdecoxib seem to be less useful for rheumatology patients, and many patients were disappointed when rofecoxib was removed from the market.

4. What recommendations do you hope that MARC will make?

• That COX-2 inhibitors remain on the market with the proviso that they are only prescribed to low cardiovascular risk patients for whom the benefits outweigh the risks, and when this benefit cannot be obtained through other sources.

• The patient needs to have an active role in the risk:benefit assessment. The risks need to be calculated on an individual basis, and the patient fully informed.
• The education of prescribers is critical in order to minimise risk.
• Access to prescribing is an issue e.g. Special Authority vs. GP prescribing.

3.1.5.3

Peri-operative use of COX-2 inhibitors

• The members discussed anecdotal reports that COX-2 inhibitors other than parecoxib and valdecoxib are also used extensively peri-operatively.
• The members considered that the greatest risk is during cardiac and vascular surgery, and other major surgery where significant COX-2 activation may occur as a result of the surgery e.g. prolonged operative periods, or surgery where tissue hypoxia is likely to occur.
• Therefore, considering the class effect for cardiovascular risk, it is considered that all COX-2 inhibitors should be contraindicated in patients undergoing cardiac or vascular surgery, and in patients at high risk of cardiovascular disease (including patients with diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia, hypertension or smokers) who are undergoing major surgery.

3.1.5.4

Familial Adenomatous Polyposis (FAP)

• Celecoxib is the only medicine approved in New Zealand for use in the treatment of FAP.
• Members commented that, although the sponsor provided evidence from a randomised controlled trial that celecoxib use decreased the number and size of colorectal polyps, there have not been any studies undertaken to assess the clinical effect of this. It was also noted that only celecoxib at a dose of 400mg bd has been shown to be effective in patients with FAP.
• Members commented that both the APC¹⁷ and PreSAP⁶⁰ trials were undertaken in patients with Spontaneous Adenomatous Polyposis (SAP) and not patients with FAP. Patients with FAP are usually younger and hence at lower cardiovascular risk compared to patients with SAP. The lifetime risk of colorectal cancer in patients with FAP is 100% and there are very limited treatment options available, other than surgery.
• Members concluded that, in low cardiovascular risk patients with FAP, the benefits of treatment with celecoxib are likely to outweigh the risks. 

3.1.5.5

Intensive Medicines Monitoring Programme (IMMP)

• It was noted that some of the pharmaceutical companies in their submissions had expressed a willingness to co-operate with IMMP monitoring of the COX-2 inhibitors. The Head of IMMP explained that it had not yet been decided what further work would be done on the COX-2 inhibitors and that this decision was dependent on funding, as currently all IMMP resources are utilised on studies of other medicines. The MARC members agreed that further IMMP studies of COX-2 inhibitors, particularly on cohorts already collected, could be performed if the pharmaceutical companies were prepared to provide funding. It was agreed that Medsafe, the IMMP Head and the Director of the New Zealand Pharmacovigilance Centre (NZPhvC) should identify options prior to the next MARC meeting.

3.1.5.6

Concurrent use of COX-2 inhibitors with Aspirin

• The members noted that there are limited data for the effect of concurrent aspirin use on cardiovascular effects due to COX-2 inhibitors.
• The members noted there is clear evidence that the concurrent use of aspirin negates the gastrointestinal benefits of COX-2 inhibitors, and instead confers about the same level of gastrointestinal harm as a NSAID alone.
• The members noted that there are conflicting opinions regarding the concurrent use of a proton-pump inhibitors in patients at very high gastrointestinal risk (e.g. previous perforation, ulcer or bleeding) if they are using a COX-2 inhibitor, and especially if they are also being treated with aspirin.
• The Committee agreed that prescribing advice for all COX-2 inhibitors should include the statement "COX-2 inhibitors are not a substitute for cardiovascular prophylaxis because of their lack of effect on platelets; therefore, concurrent anti-platelet therapies should not be discontinued. There is no evidence that concurrent use of aspirin decreases the risk of cardiovascular adverse events associated with COX-2 inhibitors. Concurrent use of aspirin negates most of the gastrointestinal benefit associated with COX-2 inhibitors."

3.1.6

Benefit-risk of all COX-2 inhibitors

The Committee considered extensive data regarding the cardiovascular risks of the individual COX-2 inhibitors, as documented above. In addition, the MARC members discussed at length the benefits of COX-2 inhibitors for particular populations of patients. 

The Committee noted that the level of acceptable cardiovascular risk varies according to the individual patient. Unfortunately, the data currently available are insufficient to establish specific guidelines with respect to cardiovascular risk thresholds for prescribers of selective COX-2 inhibitors. The Committee agreed, however, that there are electronic tools and cardiovascular risk assessment tables, such as those in MIMS New Ethicals, available to health professionals to facilitate risk assessment.

The members discussed whether there should be a 'warning' or a 'contraindication' put in place for patients at high risk of cardiovascular events. It was agreed that, if COX-2 inhibitors were to be contraindicated in high-risk (including patients with diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia, hypertension or smokers) patients, a population of patients would potentially be excluded for whom the benefits may outweigh the risks.

Following extensive discussion the following conclusions were reached:

• The members considered that COX-2 inhibitors could be used in a patient population for whom the risk:benefit analysis is favourable, for example those who are at low absolute risk for cardiovascular disease, or in whom other therapies are contraindicated or intolerable.
• The members agreed that it is very important that COX-2 inhibitors only be used at the lowest effective dose for the shortest possible time.
• The members considered that COX-2 inhibitors should be used primarily for short-term treatment of acute conditions (up to one week). However, it was acknowledged that long-term treatment with higher doses may be required for some patient populations, especially rheumatoid arthritis. In this case, the Committee considered that it is very important that the prescriber reviews the need for treatment, and the dosage, regularly. In addition, the Committee considered that, if there is a lack of efficacy after two weeks of treatment, treatment should be withdrawn.
• The Committee agreed that the limitations of the available information pertaining to usage figures for COX-2 inhibitors had restricted their ability to accurately analyse the impact of their use in New Zealand. Usage data provides a reference point from which to compare the rate of spontaneous reporting of adverse events but it is not readily available because these medicines are not subsidised in New Zealand. The Committee agreed that it should be a requirement of the sponsors of COX-2 inhibitors that they provide usage data to Medsafe on a regular, ongoing, negotiated basis.
• The Committee agreed that, in order to decrease pressure on health professionals to prescribe, and as the benefits of COX-2 inhibitor use outweigh the risks in only carefully selected patient groups, direct-to-consumer advertising (DTCA) of COX-2 inhibitors to the general population is not appropriate.
• The MARC considered that it is imperative that the data sheets for all COX-2 inhibitors be updated to adequately address cardiovascular safety concerns.

3.1.7

Conclusions

3.1.7.1

General Conclusions

• The MARC considers that there is a biologically plausible mechanism to explain an increased risk of cardiovascular thrombo-embolic events for COX-2 inhibitors.
• There are clear signals of increased cardiovascular risk associated with the COX-2 inhibitors and therefore, until proven otherwise, the cardiovascular adverse effects are considered to be a class effect for all members of this group.
• There is evidence from a number of studies to support a class effect for hypertension, congestive heart failure and renal dysfunction for both non-selective and selective COX-1 and 2 inhibitors.
• Given the limitations of the available data, all COX-2 inhibitors should be treated comparably and any restrictions placed on the products should be similar across the range of products.
• Access to specific data on the total number of patients taking COX-2 inhibitors in New Zealand would allow the Committee to assess the impact of the use of these medicines and provide context around the number and pattern of reported adverse events, and therefore would improve the ability of the Committee to make risk-management decisions in the future.
• For many patients the cardiovascular risks of COX-2 inhibitors outweigh the potential gastrointestinal benefits.

3.1.8

Recommendations

3.1.8.1

Celecoxib (Celebrex)

• The data sheet for celecoxib should contain words to the effect that:
• Use of celecoxib is contraindicated in patients who have previously had a myocardial infarction or stroke.
• Use of celecoxib in the peri- and post-operative period is contraindicated in patients undergoing cardiac or vascular surgery.
• Use of celecoxib is contraindicated in patients at high risk of cardiovascular disease (including patients with diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia, hypertension or smokers) who are undergoing major surgery.
• Use of COX-2 inhibitors (of which celecoxib is one) has been associated with increased risk of cardiovascular adverse events (myocardial infarction and stroke) when compared to patients taking naproxen and/or other less selective NSAIDs.
• Celecoxib should not be used unless alternative therapies have been used and found to lack analgesic efficacy or to have unacceptable adverse effects in the individual patient.
• Celecoxib should be used primarily for the short-term treatment of acute conditions (up to one week). In patients for whom longer-term use may be required, treatment efficacy should be reviewed after two weeks and celecoxib withdrawn if there is a lack of therapeutic benefit. Patients on long-term treatment should be reviewed every three months with regards to risk factors and the ongoing need for treatment.
• All patients taking celecoxib should commence therapy at the lowest recommended dose, and be titrated to the lowest dose compatible with effective control of symptoms for the shortest possible period.
• Prescribers should inform the individual patient of the increased risks when prescribing celecoxib for patients at high risk of cardiovascular adverse events (including patients with diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia, hypertension or smokers).
• Celecoxib may exacerbate pre-existing hypertension, cardiac failure or oedema, and the treatment of these conditions may be compromised.
• Dosage and Administration - Management of Acute Pain and Treatment of Primary Dysmenorrhoea' should read "The recommended dose is 400mg as a single dose on the first day, followed by 200mg once daily on subsequent days. Patients may be instructed to take an additional dose of 200mg on any given day, if needed. The maximum recommended dose is 400mg per day. Celebrex can be administered up to two hours prior to minor surgery."
• The statement "Doses up to 600mg daily have been studied for these indications" should be removed from 'Dosage and Administration - Management of Acute Pain and Treatment of Primary Dysmenorrhoea'.
• Celecoxib is not a substitute for cardiovascular prophylaxis because of its lack of effect on platelets; therefore, concurrent anti-platelet therapies should not be discontinued. There is no evidence that concurrent use of aspirin decreases the risk of cardiovascular adverse events associated with COX-2 inhibitors, including celecoxib.
• Concurrent use of aspirin negates most of the gastrointestinal benefit associated with COX-2 inhibitors, including celecoxib.

•  The sponsor should be required to provide Medsafe with usage data on a regular basis as a condition of continuing Ministerial consent for distribution.
• Medsafe should be asked to explore the availability of an alternative, indication-specific, celecoxib product for the treatment of Familial Adenomatous Polyposis (FAP). If such a product is available, the indication for FAP should be removed for the Celebrex brand of celecoxib.

3.1.8.2

Etoricoxib (Arcoxia)

The data sheet for etoricoxib should contain words to the effect that:

• Use of etoricoxib is contraindicated in patients who have previously had a myocardial infarction or stroke.
• Use of etoricoxib in the peri- and post-operative period is contraindicated in patients undergoing cardiac or vascular surgery.
• Use of etoricoxib is contraindicated in patients at high risk of cardiovascular disease (including patients with diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia, hypertension or smokers) who are undergoing major surgery.
• Etoricoxib is contraindicated in patients whose hypertension is not well controlled. If it is prescribed in controlled hypertension close monitoring is essential.
• All patients prescribed etoricoxib should have their blood pressure monitored regularly while taking this medication. Etoricoxib should be stopped if the patient's blood pressure is persistently raised.
• Use of COX-2 inhibitors (of which etoricoxib is one) has been associated with increased risk of cardiovascular adverse events (myocardial infarction and stroke) when compared to patients taking naproxen and/or other less selective NSAIDs.
• Etoricoxib should not be used unless alternative therapies have been used and found to lack analgesic efficacy or to have unacceptable adverse effects in the individual patient.
• Etoricoxib should be used primarily for the short-term treatment of acute conditions (up to one week). In patients for whom longer-term use may be required, treatment efficacy should be reviewed after two weeks and etoricoxib withdrawn if there is a lack of therapeutic benefit. Patients on long-term treatment should be reviewed every three months with regards to risk factors and the ongoing need for treatment.
• All patients taking etoricoxib should commence therapy at the lowest recommended dose, and be titrated to the lowest dose compatible with effective control of symptoms for the shortest possible period.
• Prescribers should inform the individual patient of the increased risk when prescribing etoricoxib for patients at high risk of cardiovascular adverse events (including patients with diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia, hypertension or smokers).
• Etoricoxib may exacerbate pre-existing hypertension, cardiac failure or oedema, and the treatment of these conditions may be compromised.
• Etoricoxib is not a substitute for cardiovascular prophylaxis because of its lack of effect on platelets; therefore, concurrent anti-platelet therapies should not be discontinued. There is no evidence that concurrent use of aspirin decreases the risk of cardiovascular adverse events associated with COX-2 inhibitors, including etoricoxib.
• Concurrent use of aspirin negates most of the gastrointestinal benefit associated with COX-2 inhibitors, including etoricoxib.

• The sponsor should be required to provide Medsafe with usage data on a regular basis as a condition of continuing Ministerial consent for distribution.
• In order to improve risk-management decision-making in the future, the sponsor of etoricoxib should be asked to supply Medsafe with further data on the safety of this product in patients with controlled hypertension.

3.1.8.3

Valdecoxib (Bextra)

The data sheet for valdecoxib should contain words to the effect that:

• Use of valdecoxib is contraindicated in patients who have previously had a myocardial infarction or stroke.
• Use of valdecoxib in the peri- and post-operative period is contraindicated in patients undergoing cardiac or vascular surgery.
• Use of valdecoxib is contraindicated in patients at high risk of cardiovascular disease (including patients with diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia, hypertension or smokers) who are undergoing major surgery.
• Use of COX-2 inhibitors (of which valdecoxib is one) has been associated with increased risk of cardiovascular adverse events (myocardial infarction and stroke) when compared to patients taking naproxen and/or other less selective NSAIDs.
• Valdecoxib should not be used unless alternative therapies have been used and found to lack analgesic efficacy or to have unacceptable adverse effects in the individual patient.
• Valdecoxib should be used primarily for the short-term treatment of acute conditions (up to one week). In patients for whom longer-term use may be required, treatment efficacy should be reviewed after two weeks and valdecoxib withdrawn if there is a lack of therapeutic benefit. Patients on long-term treatment should be reviewed every three months with regards to risk factors and the ongoing need for treatment.
• All patients taking valdecoxib should commence therapy at the lowest recommended dose, and be titrated to the lowest dose compatible with effective control of symptoms for the shortest possible period.
• Prescribers should inform the individual patient of the increased risks when prescribing valdecoxib for patients at high risk of cardiovascular adverse events (including patients with diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia, hypertension or smokers).
• Valdecoxib may exacerbate pre-existing hypertension, cardiac failure or oedema, and the treatment of these conditions may be compromised.
• Valdecoxib is not a substitute for cardiovascular prophylaxis because of its lack of effect on platelets; therefore, concurrent anti-platelet therapies should not be discontinued. There is no evidence that concurrent use of aspirin decreases the risk of cardiovascular adverse events associated with COX-2 inhibitors, including valdecoxib.
• Concurrent use of aspirin negates most of the gastrointestinal benefit associated with COX-2 inhibitors, including valdecoxib.

• The sponsor should be required to provide Medsafe with usage data on a regular basis as a condition of continuing Ministerial consent for distribution.

[Post-script: The Minister of Health’s Delegate did not accept the above MARC recommendations for valdecoxib because subsequent to the MARC meeting, following a request from Medsafe on 8 April 2005, Pfizer voluntarily withdrew valdecoxib (Bextra) from the New Zealand market.  This followed similar withdrawals in the US, Europe and Canada, due to increasing concerns about the risks of serious and potentially life-threatening cutaneous adverse reactions, as well as concerns regarding the cardiovascular safety of valdecoxib.  Consequently, the MARC recommendations with respect to valdecoxib have been overtaken by events and a further safety review is now required.  Valdecoxib will be discussed again at the June 2005 MARC meeting.]

3.1.8.4

Parecoxib (Dynastat)

The data sheet for parecoxib should contain words to the effect that:

• Use of parecoxib is contraindicated in patients who have previously had a myocardial infarction or stroke.
• Use of parecoxib is contraindicated in patients undergoing cardiac or vascular surgery.
• Use of parecoxib is contraindicated in patients at high risk of cardiovascular disease (including patients with diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia, hypertension or smokers) who are undergoing major surgery.
• Use of COX-2 inhibitors (of which parecoxib is one) has been associated with increased risk of cardiovascular adverse events (myocardial infarction and stroke) when compared to patients taking naproxen and/or other less selective NSAIDs.
• Parecoxib should only be used where other analgesics are not acceptable due to contraindications, intolerability or individual patient risk factors.
• Prescribers should inform the individual patient of the increased risks when prescribing parecoxib for patients at high risk of cardiovascular adverse events (including patients with diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia, hypertension or smokers).
• Parecoxib may exacerbate pre-existing hypertension, cardiac failure or oedema, and the treatment of these conditions may be compromised.
• Parecoxib is not a substitute for cardiovascular prophylaxis because of its lack of effect on platelets; therefore, concurrent anti-platelet therapies should not be discontinued. There is no evidence that concurrent use of aspirin decreases the risk of cardiovascular adverse events associated with COX-2 inhibitors, including parecoxib.
• Concurrent use of aspirin negates most of the gastrointestinal benefit associated with COX-2 inhibitors, including parecoxib.

In addition:

• The sponsor should be required to provide Medsafe with usage data on a regular basis as a condition of continuing Ministerial consent for distribution.

3.1.8.5

Lumiracoxib (Prexige)

The data sheet for lumiracoxib should contain words to the effect that:

• Use of lumiracoxib is contraindicated in patients who have previously had a myocardial infarction or stroke.
• Use of lumiracoxib in the peri- and post-operative period is contraindicated in patients undergoing cardiac or vascular surgery.
• Use of lumiracoxib is contraindicated in patients at high risk of cardiovascular disease (including patients with diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia, hypertension or smokers) who are undergoing major surgery.
• Use of COX-2 inhibitors (of which lumiracoxib is one) has been associated with increased risk of cardiovascular adverse events (myocardial infarction and stroke) when compared to patients taking naproxen and/or other less selective NSAIDs.
• Lumiracoxib should not be used unless alternative therapies have been used and found to lack analgesic efficacy or to have unacceptable adverse effects in the individual patient.
• Lumiracoxib should be used primarily for the short-term treatment of acute conditions (up to one week). In patients for whom longer-term use may be required, treatment efficacy should be reviewed after two weeks and lumiracoxib withdrawn if there is a lack of therapeutic benefit. Patients on long-term treatment should be reviewed every three months with regards to risk factors and the ongoing need for treatment.
• All patients taking lumiracoxib should commence therapy at the lowest recommended dose, and be titrated to the lowest dose compatible with effective control of symptoms for the shortest possible period.
• Prescribers should inform the individual patient of the increased risks when prescribing lumiracoxib for patients at high risk of cardiovascular adverse events (including patients with diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia, hypertension or smokers).
• Lumiracoxib may exacerbate pre-existing hypertension, cardiac failure or oedema, and the treatment of these conditions may be compromised.
• Lumiracoxib is not a substitute for cardiovascular prophylaxis because of its lack of effect on platelets; therefore, concurrent anti-platelet therapies should not be discontinued. There is no evidence that concurrent use of aspirin decreases the risk of cardiovascular adverse events associated with COX-2 inhibitors, including lumiracoxib.
• Concurrent use of aspirin negates most of the gastrointestinal benefit associated with COX-2 inhibitors, including lumiracoxib.

• The sponsor should be required to provide Medsafe with usage data on a regular basis as a condition of continuing Ministerial consent for distribution.

The data sheet for meloxicam should contain words to the effect that:

• Use of meloxicam is contraindicated in patients who have previously had a myocardial infarction or stroke.
• Use of meloxicam in the peri- and post-operative period is contraindicated in patients undergoing cardiac or vascular surgery.
• Use of meloxicam is contraindicated in patients at high risk of cardiovascular disease (including patients with diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia, hypertension or smokers) who are undergoing major surgery.
• Use of COX-2 inhibitors (of which meloxicam is one) has been associated with increased risk of cardiovascular adverse events (myocardial infarction and stroke) when compared to patients taking naproxen and/or other less selective NSAIDs.
• Meloxicam should not be used unless alternative therapies have been used and found to lack analgesic efficacy or to have unacceptable adverse effects in the individual patient.
• In patients for whom long-term use of meloxicam may be required, treatment efficacy should be reviewed after two weeks and meloxicam withdrawn if there is a lack of therapeutic benefit. Patients on long-term treatment should be reviewed every three months with regards to risk factors and the ongoing need for treatment.
• All patients taking meloxicam should commence therapy at the lowest recommended dose, and be titrated to the lowest dose compatible with effective control of symptoms for the shortest possible period.
• Prescribers should inform the individual patient of the increased risks when prescribing meloxicam for patients at high risk of cardiovascular adverse events (including patients with diabetes, ischaemic heart disease, cardiac failure, hyperlipidaemia, hypertension or smokers).
• Meloxicam may exacerbate pre-existing hypertension, cardiac failure or oedema, and the treatment of these conditions may be compromised.
• Meloxicam is not a substitute for cardiovascular prophylaxis because of its lack of effect on platelets; therefore, concurrent anti-platelet therapies should not be discontinued. There is no evidence that concurrent use of aspirin decreases the risk of cardiovascular adverse events associated with COX-2 inhibitors, including meloxicam.
• Concurrent use of aspirin negates most of the gastrointestinal benefit associated with COX-2 inhibitors, including meloxicam.

• The sponsor should be required to provide Medsafe with usage data on a regular basis as a condition of continuing Ministerial consent for distribution.

• Medsafe should seek assurance from the sponsor of rofecoxib that, at such a time as the sponsor intends to re-introduce rofecoxib into the New Zealand market, the MARC is given the opportunity to review the available data and to make recommendations. If this does not occur, the MARC recommends that the approval of rofecoxib be suspended, and the product will only be allowed back on the market after it has satisfied both the MARC and the Medicines Assessment Advisory Committee (MAAC) of it's cardiovascular safety.

• In order to improve risk-management decision-making in the future, the sponsors of all COX-2 inhibitors should be asked to supply Medsafe with further data on the gastrointestinal effects and cardiovascular risks of their products, as well as any adverse events that come to their attention, as a condition of continuing Ministerial consent for distribution.
• The MARC recommends that Medsafe and the New Zealand Pharmacovigilance Centre (NZPhvC) should keep a watching brief on the COX-2 inhibitor adverse drug reactions, particularly those known to occur with the conventional non-steroidal anti-inflammatory drugs (NSAIDs), such as hypertension, cardiac failure, renal dysfunction and allergic reactions.
• As further studies into the cardiovascular adverse effects of COX-2 inhibitors would be of benefit, Medsafe should investigate if sponsors can be required to fund Intensive Medicines Monitoring Programme (IMMP) studies. Medsafe should contact the New Zealand Society of Anaesthetists and the Royal Australasian College of Surgeons advising them that it is crucial that a full risk:benefit analysis is performed for all patients, and the patient fully informed before administering a COX-2 inhibitor in the peri-operative period.
• In order to decrease pressure on prescribers, and as the benefits of COX-2 inhibitor use outweigh the risks in only carefully selected patient groups, it is the MARC's opinion that direct-to-consumer advertising (DTCA) of the COX-2 inhibitor class is not appropriate and therefore should cease if these products are to remain available in New Zealand. 

[Post-script: The Minister of Health’s Delegate did not accept the MARC recommendation regarding DTCA.  The justification being that:

• the Ministry of Health is reviewing DTCA as part of the programme for establishment of the Joint Agency;
• the additional warnings on products and the release of advice to the public is sufficient to protect the public health in the interim; and
• a voluntary moratorium on DTCA and advertising to health professionals for the COX-2 inhibitor class of medicines has been in place since requested by Medsafe in December 2004;
• the voluntary moratorium is more extensive than that requested by the MARC.

The Minister’s Delegate instead agreed that, in the first instance, the intent of the MARC could be bought into effect by asking the sponsors of the COX-2 inhibitors to continue their voluntary moratorium on DTCA and professional advertising until further notice.]