|
Oct. 8, 2004 — In the wake of Merck's voluntary, worldwide
withdrawal of rofecoxib (Vioxx) on Sept. 30, concerns are surfacing
over whether the adverse cardiovascular risk documented with rofecoxib
may be a class effect extending to the other cyclooxygenase-2 (COX-2)
inhibitors. The U.S. Food and Drug Administration (FDA) issued a
public health advisory to inform patients about the withdrawal of
rofecoxib, and Acting Commissioner Lester M. Crawford announced
the FDA's intention to closely monitor other COX-2 inhibitors for
similar adverse events.
"Although the risk that an individual patient would have a
heart attack or stroke related to Vioxx is very small, the study
that was halted suggests that, overall, patients taking the drug
chronically face twice the risk of a heart attack compared to patients
receiving a placebo," Dr. Crawford said in a news release.
"All of the [nonsteroidal anti-inflammatory drugs (NSAIDs)]
have risks when taken chronically, especially of gastrointestinal
bleeding, but also liver and kidney toxicity. They should only be
used continuously under the supervision of a physician."
Mounting Data
Three-year data from the Adenomatous Polyp Prevention on VIOXX
(APPROVe) study, a multicenter, prospective, randomized, double-blind
trial, documented an increased relative risk for confirmed cardiovascular
events, including myocardial infarction and stroke, for rofecoxib
compared with placebo.
In APPROVe, which was designed to evaluate the efficacy of rofecoxib,
25 mg, in preventing recurrence of colorectal polyps in 2,600 patients
with a history of colorectal adenomas, the increased cardiovascular
risk began after 18 months of treatment with rofecoxib and persisted.
At three years, cumulative incidence of cardiovascular events was
7.5 per 1,000 patients receiving placebo compared with 15 per 1,000
patients receiving rofecoxib.
"Although we believe it would have been possible to continue
to market Vioxx with labeling that would incorporate these new data,
given the availability of alternative therapies and the questions
raised by the data, we concluded that a voluntary withdrawal is
the responsible course to take," said Raymond V. Gilmartin,
chairman, president, and chief executive officer of Merck.
The FDA approved rofecoxib in 1999 for pain and inflammation caused
by osteoarthritis, for acute pain in adults, and for menstrual pain,
and later approved it for symptoms of rheumatoid arthritis in adults
and children. Compared with nonselective NSAIDs that block both
COX-1 and COX-2, the purported advantage of rofecoxib and other
COX-2 selective inhibitors was lower risk of gastrointestinal ulcers
and bleeding, because the COX-1 enzyme helps protect the stomach
lining from acid.
In June 2000, the Vioxx Gastrointestinal Outcomes Research (VIGOR)
safety study showed an increased risk of serious cardiovascular
events. Myocardial infarction was five times higher with rofecoxib
(0.5%) than with naproxen (0.1%), and risks of stroke, venous thrombosis,
and hypertension were also significantly increased. Although Merck
argued that this difference might reflect a cardioprotective effect
of naproxen, epidemiologic studies to test that hypothesis have
proved inconclusive.
Based on VIGOR and other controlled trials, the FDA changed rofecoxib
labeling in April 2002 to include information about increased cardiovascular
risk. Safety labels notwithstanding, rofecoxib became one of the
most widely used drugs ever to be withdrawn from the market. Two
million people worldwide were taking rofecoxib at the time of the
withdrawal, resulting in $2.5 billion in rofecoxib sales last year,
and about 80 million people have taken the drug since it was first
marketed.
But now that rofecoxib is withdrawn, will use of other COX-2 inhibitors
increase, given combined sales of rofecoxib, celecoxib (Celebrex),
and valdecoxib (Bextra) — the two other approved COX-2 inhibitors,
both Pfizer products — exceeding $6 billion worldwide last
year? Or will storms brewing over rofecoxib cast doubt on the whole
class, including drugs still in the pipeline, such as Merck's etoricoxib
(Arcoxia) and lumiracoxib (Prexige) from Novartis?
"There is really a trend that all COX-2 inhibitors may have
increased risk of cardiovascular diseases following long-term use
in chronic arthritis patients, based on some meta-analyses,"
Li-Chia Chen, PhD, from the School of Pharmacy and Pharmaceutical
Sciences at the University of Manchester, U.K., told Medscape. "However,
the long-term use evidence hasn't been finalized." She added
that the U.K.'s National Institute for Clinical Excellence will
review the evidence next week, and that the guidelines for using
COX-2 inhibitors in chronic arthritis will soon be updated.
Potential Mechanisms for Cardiovascular Risk
Mechanistically speaking, there are reasons why inhibiting COX-1
and COX-2 might affect cardiovascular risk. Because COX-1 helps
promote thrombosis and COX-2 helps inhibit it, blocking COX-2 but
not COX-1 could theoretically increase the risk of myocardial infarction
and other thrombotic events. On the other hand, inflammation has
also been implicated in cardiovascular events, so controlling inflammation
via COX-2 inhibition could conceivably be protective.
In a perspective released early by the New England Journal of Medicine,
Garret A. FitzGerald, MD, a professor and chair of pharmacology
at the University of Pennsylvania School of Medicine in Philadelphia,
suggests that depression of prostaglandin I2 formation by coxibs
might be expected to elevate blood pressure, accelerate atherogenesis,
and increase the thrombotic response to rupture of an atherosclerotic
plaque. In patients at higher cardiovascular risk, coxibs would
be more likely to predispose to a clinical event early in the course
of treatment.
"The evidence for hazard from Vioxx is much stronger than
for other members of the class, [but] there is a clear mechanistic
explanation for what we saw in APPROVe and it would apply across
the class," Dr. FitzGerald told Medscape. "Although the
studies to date have been designed in a way that minimizes the likelihood
of detecting a cardivascular signal, there have been suggestions
consistent with that possibllity with all members of the class.
The outcome of APPROVe shifts the burden of responsibility to those
who claim that the results with Vioxx were due to some off target
COX-2 independent effect."
In a rabbit model (Proc Nat Acad Sci., Aug. 29, 2000), damage after
temporary coronary artery ligation was increased by treatment with
celecoxib, which completely blocked the cardioprotective effects
of the COX-2 enzyme. The authors concluded that the COX-2 enzyme
is a "cardioprotective protein," suggesting that COX-2
inhibitors may neutralize these protective effects.
Safety of Celecoxib
Although the Celecoxib Long Term Arthritis Safety Study (CLASS)
trial did not show a significantly increased rate of myocardial
infarction with celecoxib compared with the nonspecific NSAIDs ibuprofen
or diclofenac, the incidence of adverse events related to cardiac
ischemia was higher with celecoxib, especially in patients not taking
aspirin. In that group, the rate of myocardial infarction was 0.2%
with celecoxib and 0.1% with the other two drugs. For all patients,
there was a higher incidence of atrial fibrillation in the celecoxib
group than in patients taking ibuprofen or diclofenac, which was
more pronounced in the group not taking aspirin.
"There was some increase in cardiovascular events sufficient
to trigger some worried comments by the FDA physician who reviewed
the [CLASS] data," Sidney M. Wolfe, MD, director of Public
Citizen's Health Research Group, told Medscape.
Dr. FitzGerald pointed out that the original report of the CLASS
trial suggested a more favorable gastrointestinal adverse effect
profile for celecoxib than for the other NSAIDs with no increase
in cardiovascular risk. However, this report contained only six
months of data from a one-year study, and the full data showed no
difference between celecoxib and the traditional NSAIDs on predefined
gastrointestinal end points.
In a statement on Oct. 1, Pfizer concluded that based on long-term
clinical trials in more than 6,000 patients, celecoxib did not increase
the risk of myocardial infarction or stroke in patients with arthritis
and pain, even at higher-than-recommended doses.
Pfizer vice president Mitch Gandelman went on to suggest that the
company would investigate whether celecoxib might reduce cardiovascular
risk, according to a New York Times story on Oct. 4. In support
of this hypothesis, he mentioned two small-scale university studies
but acknowledged that the evidence was "scant and inconclusive."
As of press time, Mr. Gandelman and Pfizer representatives had not
returned Medscape's calls requesting comments.
Safety of Valdecoxib
Pfizer's drug valdecoxib (Bextra) is also "under a cloud because
another FDA physician reviewing the data on that drug concluded
there may be an increased cardiovascular risk," said Dr. Wolfe,
who is also an adjunct professor of internal medicine at Case Western
Reserve University in Cleveland, Ohio. "I think it is very
likely that the effect of increasing cardiovascular risk is a class
effect."
According to agency files obtained in a lawsuit by Public Citizen,
an unredacted review by the FDA Medical Officer recommended nonapproval
for valdecoxib for acute pain, citing the Coronary Artery Bypass
Graft (CABG) Surgery study 035. Even though the entire study population
received prophylactic low-dose aspirin, this trial showed an excess
of serious cardiovascular thromboembolic adverse events including
death in association with the use of valdecoxib, 40 mg twice daily,
when added to ad lib parenteral narcotic analgesia. In other databases,
including formal safety Trials 47 and 62, valdecoxib at doses higher
than 20 mg per day was associated with a greater incidence of edema
and hypertension than were ibuprofen, naproxen, and diclofenac.
Safety of Lumiracoxib
In the Therapeutic Arthritis Research and Gastrointestinal Event
Trial (TARGET) comparing lumiracoxib with naproxen or ibuprofen,
serious gastrointestinal events were significantly lower in the
lumiracoxib group, but only in patients not taking aspirin. Although
the trial lacked power to detect a difference in the rates of cardiovascular
events in nonaspirin users, there was a nonsignificant increase
in the lumiracoxib group (0.26 vs 0.18 per 100 patient-years; hazard
ratio, 1.47).
The conundrum over COX-2 inhibitors may have a chilling effect
on drugs awaiting approval, which is justified if increased cardiovascular
risk is a class effect, but unfortunate if the risk is exclusive
to rofecoxib. The FDA and other regulatory agencies might ask for
greater quantities and longer duration of safety data before approval.
Although lumiracoxib is already approved in the U.K., the FDA rejected
Novartis's application last year, requesting more efficacy data.
In a trial of 18,000 patients, lumiracoxib reduced the risk of gastrointestinal
adverse effects without increasing the risk of myocardial infarction
compared with ibuprofen and naproxen. However, in a commentary in
the Aug. 21 issue of The Lancet, Eric J. Topol, MD, from the Cleveland
Clinic Foundation in Ohio, noted that the trial excluded many people
with coronary artery disease and demonstrated potential problems
with liver toxicity.
Risk-Benefit Data
After analyzing benefit-to-risk ratio, some experts have concluded
that even a small increase in cardiovascular risk for the coxibs
may not justify their use, given limited evidence of greater efficacy
or lower gastrointestinal toxicity than other NSAIDs. In a recent
cost-effectiveness analysis by Rachel Elliott, BPharm, MRPharmS,
PhD, nonselective NSAIDs plus proton-pump inhibitors or misoprostol
were more cost-effective than COX-2 inhibitors for preventing severe
gastrointestinal bleeding, according to Dr. Chen.
"One has to wonder, what is the point of taking [coxibs],
if they don't have any advantage in terms of gastrointestinal toxicity
over the older NSAIDs," Dr. Wolfe said. "They are no more
effective than the older NSAIDs. Unlike the older NSAIDs, there
is some concern about increased cardiovascular risk."
Dr. Chen and colleagues conducted a systematic review (J Clin Pharm
Ther., June 2004;29[3]:215-229) of the analgesic efficacy and tolerability
of COX-2 inhibitors for postoperative pain control in 18 randomized
controlled trials enrolling a total of 2,783 patients. Relief of
orthopaedic pain was no different with rofecoxib 50 mg or naproxen
550 mg. Although the adverse effects of single-dose COX-2 inhibitors
were generally mild and less than nonselective NSAIDs, there was
no significant difference. Overall, the investigators concluded
that the analgesic efficacy and tolerability of single-dose COX-2
inhibitors were better than for opioid-containing analgesics and
similar to nonselective NSAIDs, but they recommended further studies
to examine the efficacy and tolerability of COX-2 inhibitors.
"Since the launch of COX-2 inhibitors, debates over their
benefits (ie, decreasing severe gastrointestinal adverse events)
and therapeutic roles is still going on. The trade-off between their
risk and benefit is still not clear," Dr. Chen said. "The
'class effect' of cardiovascular diseases is one of the concerns
with other COX-2 inhibitors.... We need more head-to-head comparisons,
long-term follow-up, and patient-centered resources use data to
answer these questions."
According to Dr. Wolfe, the COX-2 enzyme has important functions
throughout the body, including bone healing, repair of tendon rupture,
circulation to the heart, and other protective and restorative roles,
especially in emergency situations.
"The body needs to be able to produce [the COX-2 enzyme] and
use it as part of the healing process," Dr. Wolfe said. "By
inhibiting it as the COX-2 inhibitors do, more so than the older
NSAIDs, you're asking for trouble.... I think the safe way to go
is really not to use any of the drugs in this class, even the ones
left after Vioxx is coming off the market."
In response to physicians who argue that the coxibs may be more
effective in a subgroup of patients refractory to other drugs, Dr.
Wolfe counters that trials and clinical experience with older NSAIDs
may not have adequately tested a range of doses that might be equivalent
in terms of pain relief to doses that are being used for the coxibs.
"I don't think there's really any evidence at all from the
standpoint of effectiveness in pain relief in treating rheumatoid
arthritis or osteoarthritis that these COX-2 inhibitors are working
on people who have never responded to the other [NSAIDs],"
Dr. Wolfe said. His group is considering whether the evidence suggests
that the FDA should issue a box warning for other coxibs, or even
remove them from the market.
Since the withdrawal of rofecoxib, class action lawsuits have emerged
in Canada and in the U.S., and both Merck and the FDA have been
criticized for their handling of the situation. In an early-release
perspective in the New England Journal of Medicine, Dr. Topol suggests
that the debacle could have been prevented had many early warning
signs been heeded, and that neither Merck nor the FDA fulfilled
their responsibilities to the public.
Based on all the available data on rofecoxib and celecoxib available
as of Aug. 22, 2001, Dr. Topol and colleagues concluded that there
was a clear-cut excess number of myocardial infarctions associated
with rofecoxib, and an increase with celecoxib not meeting statistical
significance, mandating a trial specifically to assess cardiovascular
risk and benefit of these agents in patients with established coronary
artery disease.
Long-term Issues
Whether the damage from rofecoxib to the COX-2 class is over is
still unclear. Dr. FitzGerald advocates long-term follow-up of patients
in the APPROVe study to ascertain long-term risk.
"Many effects may wear off very quickly, but if the drug induced
hardening of the arteries, that would take longer to dissipate,"
Dr. FitzGerald said. "The best people to advise us now are
the FDA. They are an unbiased source and they have access to more
information than any investigator or any individual company. The
question is whether they should offer a warning to patients at high
cardiovascular risk concerning all members of the class until we
have more information and whether they can give us guidance on duration
of therapy in other patients."
Dr. Topol calls for a full congressional review of the rofecoxib
situation, claiming that Merck valued sales over safety, and that
the FDA's passive position of waiting for data to accrue was not
acceptable. Merck issued a response calling Dr. Topol's commentary
"flawed in many important respects" and detailing the
history of Merck's interaction with the FDA over rofecoxib.
After the VIGOR study, Dr. Wolfe suggested that the FDA could have
forced Merck to at minimum put a box warning on rofecoxib and recommend
that it should be used as a last-choice drug. "My guess is
that if the FDA had forced the company to do that, the drug would
not have been used anywhere near as much, and even though it ultimately
came off the market last week, the number of people damaged would
have been reduced enormously," he said.
Editor's Note: Dr. FitzGerald reports having received consulting
fees from NiCox and Merck and research support from Merck.
N Engl J Med. Published online Oct. 6, 2004.
|
Message
Boards
