NEWS - Celebrex has the same CV risk as Vioxx, says New Zealand safety monitoring study

[Guest guest]

Celecoxib has same CV risk as rofecoxib, says New Zealand safety monitoring

study

Rheumawire

April 15, 2005

Zosia Chustecka

Dunedin, New Zealand - A study conduced by the New Zealand Intensive

Medicine Monitoring Programme (IMMP) has found that patients taking

celecoxib (Celebrex, Pfizer) had a similar number of thrombotic

cardiovascular events (eg, myocardial infarctions and stroke) to those

taking rofecoxib (Vioxx, Merck & Co), which was withdrawn from the market

because of concerns over cardiovascular (CV) safety [1].

The results come from an interim analysis of postmarketing data on

approximately 11 000 patients followed for nearly 4 years and are published

in the May 2005 issue of Drug Safety. The authors conclude that " in

'real-life' postmarketing use in New Zealand, there is no significant

difference in cardiovascular risk for patients taking celecoxib compared

with those taking rofecoxib. " They add, " However, the number of events in

the followed-up subgroup was small and the confidence intervals for the

hazard ratio were wide. "

" It was important to perform this interim analysis at this time, as the

withdrawal of rofecoxib sparked worldwide concerns about the safety of

celecoxib, " says Dr Mira on-Woolrych (University of Otago, Dunedin,

New Zealand), head of the IMMP and principal investigator of the study.

Important contribution to growing evidence?

Drug Safety editor Rosie Stather says these results are an important

contribution to the growing international evidence on the safety of the

COX-2 inhibitors. The lack of prospective long-term data and head-to-head

comparisons has made analyzing the problem of CV risk " very difficult, " she

says. In a press release issued by the University of Otago publicizing the

analysis, Stather comments: " That is why this IMMP study is so importantit

is a postmarketing study in a 'real-life' setting, where there are none of

the exclusion criteria that apply in the highly controlled environment of a

clinical trial. "

Stather believes that this IMMP analysis " will assist healthcare decision

makers in assessing the advantages and disadvantages of celecoxib to make

recommendations regarding its use. " The New Zealand Ministry of Health is

currently reviewing all anti-inflammatory drugs and is expected to issue a

final decision before the end of this month.

Pfizer New Zealand has objected to assertions in the press release, saying

they are " incorrect, misleading, and as such extremely irresponsible. " These

results go against the large body of data showing an increased

cardiovascular risk with rofecoxib but not with celecoxib, including a

" real-life " observational study from the US FDA, as well as 3 independent

studies (Ray WA et al, DH et al, Mamdani M et al), the company says.

Pfizer medical director Dr Bill Ketelbey points out that the IMMP data are

not collected in a controlled fashion and depend on the vigilant reporting

of doctors: " It is not a clinical trial, and the results should be

interpreted with extreme caution. "

on-Woolrych comments that while the group of 11 000 patients in the

study is " clearly defined and reasonably large, it should be understood that

it may not be large enough to detect small differences in the rate of

relatively uncommon events, such as heart attacks and strokes. " Ketelbey

says that the authors themselves admit that the study lacks the statistical

power to detect a difference " between these 2 medicines that have repeatedly

been shown to have different safety profiles. "

" The inability of the IMMP to detect a known difference between these 2

medicines merely adds support to the New Zealand government's decision to

stop funding it, " Ketelbey says. This decision was made in July 2004 and

caused great outrage, with doctors from around the world writing in protest

to the Minister of Health in an open letter to the BMJ [2]. Effusive in its

praise for the system, the letter said: " Drug regulatory agencies in other

countries and professionals working on the safety of medicines have admired

and envied it, because it has so successfully identified previously

unrecognized adverse reactions and measured risks and identified risk

factors. For example, it enabled New Zealand to lead the world in taking

regulatory action over agranulocytosis due to mianserin and liver toxicity

due to nefazodone. "

The IMMP data in detail

on-Woolrych and colleagues explain that monitoring of rofecoxib and

celecoxib in New Zealand began in December 2000, shortly after both drugs

were launched, and was carried out in parallel. About 60 000 patients were

involved. At the time when rofecoxib was withdrawn (September 2004),

follow-up had been completed on about one fifth of these patients (n=11

149). Of these, 6267 were taking celecoxib and 4882 were taking rofecoxib

(19% and 18%, respectively, of the total cohorts). In the remaining

patients, follow-up was incomplete for 14 610 celecoxib patients (45% of

total) and 11 587 rofecoxib patients (44%), and there was no follow-up in a

further 11 569 celecoxib patients (36%) and 9934 rofecoxib patients (38%).

All of the patients were included in the analysis of thrombotic CV events.

Most patients on rofecoxib were taking the 25-mg/day dose (64.5%), about one

quarter were taking the lower dose of 12.5 mg/day, and 11% patients were

taking the high dose of 50 mg/day. For celecoxib, most patients were taking

200 mg/day (81.6%), with 10.9% taking 400 mg/day and 7.1% taking 100 mg/day.

Thrombotic CV events were identified from several different sources:

follow-up questionnaires, hospital admission data, spontaneous reports, and

prescription data. Events were examined by clinical assessors and coded

using terms from the IMMP dictionary, which is based on WHO Adverse Reaction

Terminology. These included myocardial infarction as well as terms that

suggested myocardial ischemia (eg, angina, unstable angina, ischemic heart

disease, and cardiac chest pain) in addition to stroke and transient

ischemic attack (TIA), although any events suggesting hemorrhagic stroke or

intracranial hemorrhage were excluded.

In the total celecoxib cohort, the 86 patients with thrombotic CV events

(0.27% of the cohort)) included 60 with myocardial infarction/ischemia (of

whom 34 died) and 26 with cerebral thrombotic events (22 strokes and 4 TIAs,

with 18 deaths). Two thirds of these patients (57/86, 66%) had documented

preexisting cardiovascular disease.

In the total rofecoxib cohort, the 58 patients with thrombotic CV events

(0.22% of the cohort) included 42 patients with myocardial

infarction/ischemia (24 deaths) and 16 with cerebral thrombotic events (11

strokes, 3 TIAs, 1 recurrent TIA, and 1 cerebral thrombosis; 8 deaths).

Again, about two thirds of these patients (37/58, 64%) had documented

preexisting cardiovascular disease.

The authors note that there were differences between the patients taking the

2 drugs. The celecoxib cohort was older, and more patients were taking the

medicine long term, taking a concomitant nonsteroidal anti-inflammatory drug

(NSAID), and had preexisting cardiovascular disease. However, they comment

that they accounted for these differences by adding several variables to the

statistical models, but after the adjustment for age, the adjustment for

other variables made little difference to the results.

" The most confidence " can be placed in the data from the cohort of patients

who had completed follow-up, the authors comment. In this group (n=11 149),

the unadjusted hazard ratio for celecoxib compared with rofecoxib was 1.07

(95% CI 0.59-1.93). After adjustment for age, this hazard ratio was 0.94

(95% CI 0.51-1.70). Further adjustment (for sex, indication for use,

concomitant NSAID use, " as-required " use, and preexisting cardiovascular

disease) resulted in only minor changes to the hazard ratio, the authors

add.

Approached by rheumawire for comment, Dr Matteson (Mayo Clinic,

Rochester, MN), who is an editorial consultant for www.jointandbone.org,

says he would describe this as " a noteworthy study " but " the high number of

patients with poor to no follow-up limits the confidence I have in the

results. And, of course, we do not know the expected rates for the CV events

in the population, so it is not possible to say on the basis of this study

whether these drugs have an increased risk of CV events, as the study was

not designed to address this issue. " Another editorial consultant, Dr

Felson (Boston University, MA), adds that the number of thrombotic CV events

is " not enough for conclusions. "

Sources

on-Woolrych M, on P, McLean R, et al Incidence

of thrombotic cardiovascular events in patients taking celecoxib compared

with those taking rofecoxib. Drug Saf 2005; 28:435-442.

Herxheimer A. Open letter to Annette King, Minister of

Health, New Zealand. BMJ. 2004 Jul 3; 329:51.

Not an MD

I'll tell you where to go!

Mayo Clinic in Rochester

http://www.mayoclinic.org/rochester

s Hopkins Medicine

http://www.hopkinsmedicine.org