Cox-2 Inhibitors: Today and Tomorrow
ROBERT W. McMURRAY, MD; KENNETH J. HARDY, MD
ABSTRACT: The elucidation of inducible cyclooxygen-
ase (Cox-2) dependent inflammatory pathways led to the
development of specific Cox-2 inhibitors, the coxibs.
These agents include the currently available celecoxib
and rofecoxib and such second-generation agents as
parecoxib, valdecoxib, and etoricoxib. The therapeutic
advantage of coxibs is founded primarily in their lack of
significant gastrointestinal (GI) side effects. Clinical trials
have demonstrated the efficacy of coxibs to be com-
pletely comparable with traditional nonsteroidal anti-
inflammatory drugs (NSAIDs), and pharmacoeconomics
suggest favorable cost/benefit ratios with these agents
compared with traditional NSAIDs, related to their re-
duced GI complication profiles and lower indirect costs
associated with disability. Although several clinical
questions remain (eg, use with low-dose aspirin, risk of
S everal recent articles1–5 have documented inves-
tigational and clinical experience during the
dawning of a new era: the design and implementa-
tion of cyclooxygenase-2 (Cox-2) specific inhibitors in
the treatment of rheumatic disease. Over the past
24 months, the introduction, marketing, utilization,
and postmarketing experience of the first 2 Cox-2
specific inhibitors, celecoxib and rofecoxib, have pro-
duced dramatic changes, challenges, and insights
into the treatment of rheumatic diseases. Further-
more, a number of additional Cox-2 specific drugs
are currently under development, implying that
even greater utilization of this class of medications
can be expected. Herein, we review available clinical
evidence regarding the utilization, efficacy, and
safety profiles of celecoxib and rofecoxib and intro-
duce preliminary data on 3 new Cox-2 inhibitors:
parecoxib, valdecoxib, and etoricoxib.
The novelty of Cox-2 inhibitor therapy derives
largely from the extremely brief historical time span
(about a decade) between the fundamental identifi-
cation of an important new molecular pathway me-
From the Division of Rheumatology/Molecular Immunology,
Department of Internal Medicine, University of Mississippi Med-
ical Center, and the Rheumatology Section, G.V. (Sonny) Mont-
gomery VA Hospital, Jackson, Mississippi.
Correspondence: Robert W. McMurray, M.D., Division of Rheuma-
tology and Molecular Immunology, L525 Clinical Sciences Building,
University of Mississippi Medical Center, 2500 North State Street,
Jackson, MS 39216 (E-mail: robert.mcmurray@med.va.gov).
THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
thrombosis, myocardial infarction, edema, and hyper-
tension), the emergence and clinical utility of coxibs is
likely to continue on the basis of their efficacy and
relative GI safety advantage. Although newer, more
specific Cox-2 inhibitors may alter the choice, it is likely
that this class of anti-inflammatories will become (if they
have not already) the drugs of first choice in the treat-
ment of acute pain, chronic pain, and most rheumatic
conditions in the 21st century. In addition to the treat-
ment of rheumatic conditions, it is possible that coxibs
will also be of clinical utility in protection against ma-
lignant transformation and Alzheimer disease. KEY IN-
DEXING TERMS: Cox-2 inhibitors; Coxibs; Celecoxib;
Rofecoxib; Parecoxib; Valdecoxib; Etoricoxib. [Am J
Med Sci 2002;323(4):181–189.]
diating inflammation and the development of potent
pharmacological agents capable of specifically inhib-
iting that pathway. Until 1990, conventional scien-
tific theory taught that during local and/or systemic
inflammatory responses, intracellular arachidonic
acid liberated from cell lipid membranes by a variety
of stimuli was converted into a common prostaglan-
din precursor under the influence of a single en-
zyme, cyclooxygenase. Around 1990, several inves-
tigators discovered 2 closely related forms of the
cyclooxygenase enzyme, cyclooxygenase 1 (Cox-1)
and cyclooxygenase 2 (Cox-2). Both of these isoen-
zymes synthesize arachidonic acid into the common
prostaglandin precursor, and subsequent modifica-
tion through the cell- and stimulus-specific syn-
thetases leads to local prostaglandin release with
variable functional consequences. Cox-1 is the pre-
dominant constitutive form of the enzyme and is
expressed widely throughout the body, providing
homeostatic functions such as gastric mucosal main-
tenance, platelet aggregation, fluid and electrolyte
balance, and, under certain circumstances, renal
circulation. To some extent, Cox-2 is also expressed
constitutively in several tissues, including the brain,
reproductive organs, kidney, and placenta during
late gestation. However, Cox-2 is usually absent in
most tissues but is rapidly expressed in response to
a number of local inflammatory stimuli. Its expres-
sion may occur in many tissues and in several im-
portant pathophysiological states, including acute
and chronic inflammation, hyperalgesia, various
181
Cox-2 Inhibitors
models of arthritis and glomerulonephritis, rheuma-
toid arthritis, angiogenesis, bone resorption, gastric
ulceration, colon cancer, and Alzheimer disease. The
immediate effects of Cox-2 induction include en-
hanced mediation of pain and inflammation, sys-
temic pyrogenicity, and local vasodilatation and/or
vasoconstriction.1,3
Pharmacology
The discovery and characterization of Cox-2 led to
the design of specific Cox-2 inhibitors (coxibs). Non-
selective nonsteroidal anti-inflammatory drugs
(NSAIDs) inhibit both Cox-1 and Cox-2 isoforms to a
similar extent at therapeutic concentrations. The
theoretical advantage of the coxibs depends on their
ability to variably inhibit only Cox-2, thus providing
anti-inflammatory and analgesic effects while avoid-
ing the gastrointestinal toxicity and platelet inhibi-
tion associated with Cox-1 inhibition. Two currently
available Cox-2 specific inhibitors, celecoxib and ro-
fecoxib, are 7.5 and 35 times more selective, respec-
tively, for Cox-2 than for Cox-1.6 Examination of
steady-state inhibitory activity of rofecoxib on Cox-2
versus Cox-1 was compared with that of other com-
monly used NSAIDs in 76 healthy volunteers.7 Ro-
fecoxib (25 mg/d), meloxicam (15 mg/d), diclofenac
(50 mg 3 times/d), ibuprofen (800 mg 3 times/d), and
naproxen sodium (550 mg twice/d) were compared
with one another and with placebo. Ex vivo whole-
blood assays were used to determine effect on Cox-2
and Cox-1 activity, respectively. Mean inhibition of
Cox-2 [measured as the weighted average inhibition
of lipopolysaccharide (LPS)-induced prostaglandin
E2 generation over 8 hours on day 6 versus baseline]
was 69, 78, 94, 71, 72, and ⫺2%, for rofecoxib,
meloxicam, diclofenac, ibuprofen, naproxen, and
placebo, respectively. Corresponding values for
mean inhibition of COX-1 (measured as thrombox-
ane B2 generation in clotting whole blood) were 7,
53, 50, 89, 95, and ⫺5%. Further, rofecoxib had no
significant effect on urinary excretion of 11-dehydro-
thromboxane B2, a Cox-1-derived urinary prosta-
noid. These data and considerable other evidence
support the concept that coxibs can uniquely inhibit
Cox-2 with minimal effects on Cox-1 activity.6,7
The major initial clinical impetus for developing
these drugs derived from their Cox-2 specificity,
which would predict a lower incidence of gastroin-
testinal bleeding compared with traditional
NSAIDs. Before the development of Cox-2 specific
inhibitors, conventional NSAIDs had long been rec-
ognized as a significant source of unacceptable mor-
tality and morbidity,8 with as many as 1 in 5 persons
developing endoscopic ulcers after using NSAIDs for
more than 3 months. In America, more than 16,000
deaths per year have been attributed to the gastro-
intestinal (GI) complications from traditional
NSAIDs, especially GI bleeding, perforation, and
182
obstruction.9 Selective inhibition of Cox-2 should be
an attractive therapeutic option to traditional
agents in both long- and short-term management of
pain and inflammatory conditions such as rheuma-
toid arthritis and osteoarthritis. Clinical evidence of
substantial improvement in gastrointestinal safety
has now been well demonstrated in the setting of
Cox-2 specific inhibition. Human clinical trials (de-
scribed below) have convincingly demonstrated that
both celecoxib and rofecoxib induce significantly
fewer gastrointestinal complications compared with
nonselective NSAIDs. However, some controversy
still exists relating the use of coxibs in combination
with low-dose aspirin. The antiplatelet properties of
aspirin are well known to lower the risk of myocar-
dial infarction, stroke, and thromboembolism. Be-
cause Cox-2 inhibitors alone have no impact on
platelet function, they will offer no protection to
high-risk cardiovascular patients, and concomitant
low-dose aspirin would logically be indicated. How-
ever, the optimal dose (81 versus 325 mg), timing
(every day, every other day, etc) and/or delivery
system (uncoated, coated, chewable) for aspirin to be
used safely with coxibs in thromboprophylaxis has
not yet been determined.
In general, other well-known traditional NSAID
contraindications, such as allergy, use in late preg-
nancy, aspirin-induced asthma, congestive heart
failure, and renal dysfunction, also apply to the
Cox-2 selective inhibitors.1–5 Recent preliminary re-
ports have suggested that the coxibs may in fact be
well tolerated in aspirin-induced asthma, perhaps
promising a safe utility for their use in managing
persons allergic to NSAIDs.10 –12 At present, how-
ever, such use is not approved. Although allergic
reactions to the sulfonamide group of celecoxib are
reported infrequently,13–15 the incidence of skin rash
was more than double that of comparable NSAIDs in
a recent large clinical trial,16 and celecoxib is cur-
rently still contraindicated in patients with a known
sulfonamide allergic history17(Table 1). Many of the
relative contraindications to coxib use as a class in
congestive heart failure and renal dysfunction result
from recent studies demonstrating Cox-2 constitu-
tive expression in the kidney. The Cox-2 pathway
becomes important in regulating several local and
systemic prostaglandin-dependent renal mecha-
nisms, especially during renal dysfunction.18 –22
Even in healthy persons, rofecoxib and celecoxib
may produce small but qualitatively significant
changes in urinary prostaglandin excretion, glomer-
ular filtration rate, and sodium retention, with con-
sequences similar to those of nonselective
NSAIDs.18 –20,23,24 Thus, in both healthy and compro-
mised renal patients, drugs that specifically inhibit
Cox-2 might be expected to have many of the same
detrimental effects on renal function as nonselective
NSAIDs. It seems unlikely that available coxibs
(and probably even more specific Cox-2 inhibitor
April 2002 Volume 323 Number 4
 McMurray and Hardy

Table 1. Clinical Characteristics of Available and Investigational Cox-2 Inhibitors

Celecoxib Rofecoxib Parecoxib Valdecoxib Etoricoxib

Trade name Celebrex (Pfizer) Vioxx (Merck) Not released Bexfra (Pharmacia) Not released
(Pharmacia) (Merck)
FDA Approval OA, RA, FAP OA Pending OA, RA Pending
Familial adenomatous Acute pain/dysmenorrhea Primary dysmenorrhea
polyposis
Acute pain/dysmenorrhea
Administration Oral Oral Intravenous Oral Oral
Intramuscular
Dose 100 mg BID 12.5 or 25 mg QD, 50 mg 10 mg QD
QD (ⱕ5 days for acute 20 mg BID
pain) (dysmenorrhea)
200 mg QD or BID
400 mg BID (for FAP)
400 mg QD
Action Selective Cox-2 inhibition Specific Cox-2 inhibition Specific Cox-2 Specific Cox-2 inhibition Highly Specific
inhibition Cox-2 inhibition
Cox-2/Cox-1 Inhibition 7-fold 35-fold 30-fold 30-fold 106-fold
Metabolism Hepatic oxidation Hepatic reduction Hepatic oxidation P450
cytochrome P450 system (flavoprotein reductase) system (209
(2C9 isoenzyme) isoenzyme)
Excretion Hepatic/renal Renal Hepatic
Contraindication Sulfonamide allergy None None
Drug interactions Reported increases in Slight increases in serum Slight increases in
warfarin, cytochrome levels of warfarin, serum levels of
P450 2C9 inhibitors 2D6 methotrexate, rifampin warfarin, lithium
Cleared drugs
Common Side Dyspepsia (8.8%), rash Edema (3.7%),c Dyspepsia (7.9%),
Effect(s)a (2.2%), edema (2.1%),b hypertension nausea (7%)
hypertension (⬍2.0%)b (3.5%),c, dyspepsia (3.5%)

BID, 2 times/d; QD, 1 time/day;

a
Percentage incidence of side effects from prescribing information.
b
Prolonged use of celecoxib at 400 mg BID may lead to an increased incidence of edema (3.7%) or hypertension (2.0%).
c
Prolonged use of rofecoxib 50 mg QD may lead to an increased incidence of edema (6.9%) or hypertension (8.2%).

successors) will offer substantial renal safety bene-
fits over nonselective NSAID therapies, although
individual pharmacological differences may confer
specific advantages. A potential exception to this
may be in the early phase of autoimmune glomeru-
lonephritis (in both murine and human models),
where Cox-2 induced intrarenal inflammation and
vasoconstriction seem to be in part offset by Cox-2
inhibition.25,26 Currently, one must assume that all
NSAIDs, including Cox-2–selective inhibitors, share
a similar risk for adverse renal effects based on their
potential for inhibition of renal prostaglandin syn-
thesis. Potential differences in the relative magni-
tude of renal adverse events such as increased blood
pressure and/or peripheral edema with rofecoxib
compared with celecoxib have been reported in a
recent industry-sponsored trial.20 However, critics
have highlighted important differences in drug dos-
ing, population demographics, and pharmacody-
namics as possible explanations for the observed
differences. In another large industry-sponsored os-
teoarthritis (OA) trial, no differences in incidence of
hypertension was observed with rofecoxib compared
with traditional NSAIDS.27
Recent reports of thrombosis with celecoxib28 and
increased incidence of myocardial infarction with
rofecoxib relative to naproxen29 also confound any
clear cardiovascular superiority of coxibs compared
with nonselective NSAIDs and imply the need for
patient scrutiny and further investigation in this
area.30,31 Cox-1 and Cox-2 can variably catalyze the
formation of prothrombotic (eg, thromboxane)
and/or antithrombotic (eg, prostacyclin) prostaglan-
dins. Aspirin (an irreversible Cox-1 inhibitor), con-
ventional NSAIDs (reversible, competitive Cox-1 in-
hibitors), and coxibs (virtually noninhibitory to
Cox-1) may exhibit variable inhibitory effects on
Cox-1 mediated thromboxane generation, and per-
haps also Cox-2 mediated prostacyclin biosynthesis.
The biology of these vasoactive eicosanoids and
pharmacological effects of their differential inhibi-
tion on thromboprophylaxis and/or thrombogenicity
is incompletely understood.24,30 –32 Investigations
are needed to determine what risk, if any, of en-
hanced thrombogenicity actually exists relative to
placebo when Cox-2–specific inhibitors are used
alone in high-risk patients. At the very least, further
studies will be necessary to examine and optimize
conditions for using aspirin and/or other platelet
inhibitors concomitantly with nonspecific, selective,

THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES 183

Cox-2 Inhibitors
or specific cyclooxygenase inhibitors to optimally
balance platelet inhibition/thromboprophylaxis, va-
soactivity, and gastric preservation.
A recent meta-analysis of thrombotic events asso-
ciated with coxibs has been published and suggests
that Cox-2 inhibition may be associated with an
increased rate of cardiovascular events compared
with other NSAIDs.33 Although provocative, this
meta-analysis has several methodological difficul-
ties that restrict its interpretation and applicability
to clinical practice. These difficulties include, but
are not necessarily limited to, disparities in the
cohort used for comparisons. OA and rheumatoid
arthritis (RA) placebo cohorts were not used in the
meta-analysis; therefore, the comparison is not valid
on the basis of age, sex, and a variety of risk factors.
Moreover, the celecoxib CLASS16 and the rofecoxib
VIGOR29 trials were not designed as cardiovascular
event outcome studies and merely documented car-
diovascular events as side effects. The enrollment
and detection methodologies of a primary myocar-
dial infarction prevention trial differ markedly from
those designed to measure efficacy and GI outcomes
in Cox-2 arthritis trials. The lack of validity of this
meta-analysis seems verified by the direct compari-
son of cardiovascular event incidence of celecoxib
and rofecoxib to NSAID comparators. In those stud-
ies, there was no apparent difference in overall myo-
cardial infarction incidences between celecoxib, ro-
fecoxib, diclofenac, and ibuprofen16,29 and the
observation of a reduced myocardial infarction inci-
dence with naproxen sodium versus rofecoxib was a
novel finding.29 Although the potential for inducing
cardiovascular events in persons at high risk may
exist, this possibility must also be weighed clinically
against the well-defined higher risk and incidence of
gastrointestinal bleeding induced by nonselective
cyclooxygenase inhibition.
Although the average retail cost of the 2 available
coxibs remains high, several traditional NSAIDs are
similarly priced. Most health care economic models
comparing coxibs to nonselective NSAIDs or
NSAIDs with gastroprotective agents demonstrate a
more favorable economic benefit with coxib use,34 –38
probably because of reduced incidence of GI compli-
cations. Cox-2 specific inhibitors have a significant
potential for reducing health-care costs, primarily
through reduction in hospital expenses associated
with serious gastrointestinal ulceration, perfora-
tion, and bleeding. Reduced indirect costs through
improved disability scores and improved health-re-
lated quality of life are also predictable with the use
of Cox-2-specific inhibitors. Pharmacoeconomic com-
parison models of sodium salicylate or magnesium
trilisate (other relative gastroprotective and plate-
let-sparing NSAIDs) have not, to our knowledge,
been performed. The latter agents, although report-
edly GI-sparing, can be uniquely associated with
salicylate-specific side effects such as tinnitus. One
184
cost-effective and safe approach might be to use
nonselective NSAIDs for short-term purposes, espe-
cially in low risk and/or younger patients (ie, those
with no ongoing bleeding events, a negative GI his-
tory, etc). Over-the-counter NSAIDs could thus re-
main the principle therapy for such persons. Coxibs
on the other hand, could be prescribed acutely to
older patients and young persons at high risk for GI
bleeding, as well as to any persons having appropri-
ate indications for chronic NSAID therapy. As the
number of available Cox-2 specific drugs increases
in a competitive market, retail prices should fall.
Clinical Application
Celecoxib
Celecoxib (Celebrex; Pfizer, Inc., New York, NY)
was the first commercially available Cox-2–specific
inhibitor. It was approved for relief of the signs and
symptoms of OA and RA, as well as for treatment of
familial adenomatous polyposis (FAP).39 – 42 Cele-
coxib is a highly lipophilic drug with a large volume
of distribution (⬎ 400 L) and rapid, cytochrome
P450-dependent, hepatic inactivation (~500 mL/
min). Additional pharmacokinetic and clinical utili-
zation data are shown in Table 1. For both OA and
RA, celecoxib has been shown to be significantly
superior in efficacy to placebo and similar in efficacy
to traditional nonsteroidal anti-inflammatory drugs.
The main advantage over traditional NSAIDs is its
gastrointestinal (GI) safety. At recommended doses,
no dose-response relationship between celecoxib and
upper GI symptoms has been demonstrated. Early
randomized clinical trials demonstrated a GI safety
profile for celecoxib superior to that of traditional
NSAIDs such as naproxen, and similar to that of
placebo in patients with OA and RA.39 – 42 Recent
FDA guidelines, however, have prohibited coxib
manufacturers from removing warnings of gastric
risk from advertisements and package inserts.43
This relates, in part, to concerns over the impact of
low-dose aspirin on the gastroprotective effect oth-
erwise expected with coxib.44 A profile of renal and
cardiovascular safety with celecoxib, as well as a
demonstration of its efficacy, tolerability, and safety
in the elderly population, has been suggested in
industry-sponsored clinical trials.20 However, in
some case reports, celecoxib was associated with
gastrointestinal bleeding,44,45 hypoprothrom-
binemia, or thrombosis.28 It is not clear whether
44
such celecoxib-associated thrombosis is related to
other hypercoagulable states, such as antiphospho-
lipid antibody syndrome.31
Large multicenter trials have shown that cele-
coxib (200 or 400 mg twice/d) is as effective as
naproxen (500 mg twice/d) in patients with RA.40 A
comparative trial showed that celecoxib (200 mg
twice/d) is as effective as diclofenac SR (75 mg
twice/d) in patients with OA.42 Thus, this Cox-2
April 2002 Volume 323 Number 4

specific inhibitor provides anti-inflammatory and
analgesic efficacy equivalent to that of conventional
nonsteroidal anti-inflammatory drugs without ad-
verse gastrointestinal or platelet effects. This para-
digm, if ubiquitous among all coxibs, promises to
revolutionize the clinical care of arthritis pa-
tients.34 –36 Although pharmacologically well suited
for the management of chronic pain and inflamma-
tion, the optimal analgesic and/or anti-inflamma-
tory dose of celecoxib for acute pain has not yet been
determined; at present, acute pain indications are
not approved. Preliminary studies do suggest that
celecoxib compares favorably with narcotic analge-
sia with hydrocodone46 and that, optimally dosed,
celecoxib might prove useful in treating postopera-
tive pain.47
In terms of its GI safety profile, the most-refer-
enced celecoxib trial16 was the Celecoxib Long-term
Arthritis Safety Study (CLASS). This was a double-
blind, randomized, controlled trial involving 8059
patients with OA or RA. Patients were randomly
assigned to receive celecoxib (400 mg twice/d, 2 and
4 times the maximum recommended RA and OA
dosages, respectively; n ⫽ 3987); ibuprofen (800 mg,
3 times/d; n ⫽ 1985); or diclofenac (75 mg twice/d; n
⫽ 1996). Aspirin use for cardiovascular prophylaxis
(ⱕ325 mg/d) was permitted and used in approxi-
mately 20% of patients. End points were evaluated
for celecoxib (coxib arm) versus ibuprofen and di-
clofenac combined (NSAID arm). Although data
from only the first 6 months of the study were
published,16 the entire 13-month study was later
analyzed in detail and presented to the FDA.48,49
The 6-month data showed that for all patients in the
study, the annualized incidence of serious upper GI
ulcer complications was 0.76 versus 1.45% (P ⫽ 0.09,
celecoxib versus NSAIDs, respectively), and for
symptomatic ulcers was 2.08 versus 3.54% (P ⫽
0.02). The significant reduction in symptomatic ul-
cers with celecoxib relative to the combined compar-
ator NSAIDs was found at both 6 and 13 months.
However, the downward trend noted in serious up-
per GI complications was not statistically significant
at 6 months, and was even less apparent at the
study’s 13-month completion (P ⫽ 0.45). It was sug-
gested that aspirin use (ⱕ325 mg/d) may have con-
tributed to this finding. Indeed, in the first 6 months
of the study, those patients not taking aspirin had
an annualized incidence of serious upper GI compli-
cations that dropped significantly to 0.44 versus
1.27% (P ⫽ 0.04, celecoxib versus NSAIDs). Again,
however, that statistical benefit was lost at 13
months (P ⫽ 0.19). In persons not using aspirin,
symptomatic ulcer rates for celecoxib versus com-
bined NSAIDs dropped to 1.40 versus 2.91% (P ⫽
0.02) throughout the entire study. These data glo-
bally suggested a 20 to 50% reduction in annualized
incidence rates for all GI events with celecoxib, at
least in persons not using aspirin. The GI benefit
THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
McMurray and Hardy
was similar with celecoxib and diclofenac, both of
which are more Cox-2–selective drugs. Notably, in
patients taking either celecoxib or a comparable
NSAID plus aspirin, the annualized incidence of
serious upper GI complications and symptomatic
ulcers was considerably higher and virtually identi-
cal for celecoxib versus comparable NSAIDs. Para-
doxically, at the end of the study, there was a strong
trend demonstrating a reduction in GI events using
ibuprofen/aspirin combinations compared with ei-
ther celecoxib/aspirin or diclofenac/aspirin, the lat-
ter 2 being indistinguishable. Such observations
highlight the potential difficulties in predicting GI
outcomes when using coxib/aspirin combinations,
and point to the need for further studies. It is nota-
ble that fewer celecoxib-treated patients than
NSAID-treated patients experienced chronic GI
blood loss, GI intolerance, hepatotoxicity, or renal
toxicity. Further, no difference was noted in the
incidence of myocardial infarction between celecoxib
(0.5%) and NSAIDs (0.4%), regardless of aspirin use.
Although not statistically conclusive, the CLASS
study did suggest that celecoxib (without concomi-
tant aspirin) at dosages at least twice those recom-
mended clinically was globally associated with a
lower incidence of both symptomatic ulcers and se-
rious ulcer complications (as well as other clinically
important adverse effects), relative to standard dos-
ages of comparator NSAIDs. This suggested that in
patients not taking aspirin, there was a safety ad-
vantage with celecoxib relative to traditional
NSAIDs, with decreases in upper GI toxicity being
proportional to Cox-2 selectivity.26,48,49
An important, potential utilization of Cox-2 inhib-
itors may be in long-term prevention of colorectal
cancer, especially patients with FAP, hereditary
nonpolyposis colorectal cancer, and sporadic adeno-
mas. Preliminary studies have shown a significant
reduction in adenoma burden in familial adenoma-
tous polyposis patients who received the selective
Cox-2 inhibitor celecoxib, and FDA approval for that
indication has been granted. Obviously, design and
implementation of additional clinical trials will be of
paramount significance in validating and extending
these preliminary results.50 –52
In conclusion, celecoxib was the first coxib intro-
duced with a low–GI-side-effect profile, overall
safety at even high doses, and proven efficacy in the
treatment of OA, RA, and FAP. Concern does re-
main over its safety when used concomitantly with
aspirin and/or with warfarin (discussed below), and
the potential class-wide effect on altering coagula-
tion kinetics and renal function need further
examination.
Rofecoxib
Rofecoxib (Vioxx; Merck & Co., Inc., Whitehouse
Station, NJ) was the second coxib to become com-
mercially available. It is an oral medication cur-
185
Cox-2 Inhibitors
rently approved for treatment of acute pain, primary
dysmenorrhea, and OA. It has been demonstrated to
be efficacious in the treatment of RA29 but does not
yet carry that FDA indication. It is pharmacologi-
cally distinct from celecoxib in having rapid and
complete absorption, a smaller (~90 L) volume of
distribution (producing rapid tissue equilibrium), a
3-fold slower hepatic inactivation rate (by cyto-
chrome P450 independent pathways), and nearly a
5-fold greater Cox-2 specificity. At its recommended
acute pain dose, rofecoxib can provide anti-inflamma-
tory serum levels within 24 hours.53 Its slower inacti-
vation results in a long half-life (17 hours), permitting
once-a-day dosing for all indications. Table 1 lists ad-
ditional pharmacokinetic and clinical utilization data
regarding rofecoxib. Rofecoxib’s high Cox-2 specificity
has predicted a substantial GI safety advantage. Tri-
als comparing rofecoxib with ibuprofen, diclofenac,
and indomethacin demonstrated significantly less GI
toxicity, as measured by decreased endoscopic lesions,
lack of GI blood loss in sensitive assays, and a fre-
quency of serious adverse GI events indistinguishable
from placebo. The presence of constitutive Cox-2 in
cells not involved in inflammation, especially in the
kidney, would predict the variable occurrence of side
effects from rofecoxib common to all NSAIDs and
coxibs, namely peripheral edema and hypertension
(Table 1). Like traditional NSAIDs, these side effects
seem to be mechanism-based, potency- and dose-de-
pendent, and clinically apparent in 2 to 5% of rofecoxib
users at recommended doses.
Rofecoxib is currently FDA-approved for the treat-
ment of OA and compares favorably with diclofe-
nac.54 –56 Although not currently FDA-approved for
the treatment of RA, rofecoxib has been studied at
50 mg/d in RA treatment and has been shown to
compare favorably with naproxen.29,53 In its most
expansive GI outcome trial (VIGOR), more than
8000 RA patients were randomly assigned to receive
50 mg of rofecoxib daily (twice the maximum recom-
mended dose for OA) versus 500 mg of naproxen
twice daily, for 12 months.29 The primary end point
was confirmed clinical upper gastrointestinal events
(perforations, obstructions, GI bleeding, and symp-
tomatic ulcers), and the secondary end point was
confirmed serious gastrointestinal events (perfora-
tions, obstructions, and major bleeding). In that
study, rofecoxib and naproxen were incidentally
shown to have similar efficacy in managing rheuma-
toid arthritis, confirming the earlier report,29 and
the FDA fully concurred with the GI safety conclu-
sions exactly as published.53
The VIGOR study was statistically powered and
fully completed for both primary and secondary GI
endpoints. Rofecoxib treated RA patients had signif-
icantly fewer upper gastrointestinal events (2.1 ver-
sus 4.5%, primary endpoint, P ⫽ 0.001; 0.6 versus
1.4%, secondary endpoint, P ⫽ 0.005). Globally, the
risk of GI events in the rofecoxib arm of the study
186
was lowered more than 2-fold relative to naproxen,
even in high-risk (old age, corticosteroid use, previ-
ous GI events, etc) patients. There was nearly a
10-fold reduction in risk of adverse GI events in
persons at low risk, suggesting very little risk at all
with short-term use in low-risk patients. Neither the
rofecoxib trials nor the celecoxib trials employed
placebo arms, so GI risks relative to NSAID-un-
treated patients could not be directly ascertained in
either study. In contrast to the celecoxib trial,16
rofecoxib patients in the VIGOR trial29 were not
allowed to use aspirin despite the higher incidence
of cardiovascular disease associated with RA.57,58 It
was observed serendipitously that the incidence of
myocardial infarction was lower among patients in
the naproxen group (0.1%) than in the rofecoxib
group (0.5%), even though the overall mortality rate
and the rate of death from cardiovascular causes
were identical in both arms of the study. The differ-
ence in myocardial infarction rate between rofecoxib
and naproxen was attributed to the long-acting,
platelet inhibiting effect of naproxen, which could
theoretically provide cardioprotective benefits.
There have been no controlled studies yet examining
that hypothesis, because prophylaxis of cardiovascu-
lar events related to prostaglandin inhibition by drugs
other than salicylates has been poorly studied. How-
ever, in other large studies with patients with OA
treated with rofecoxib compared with traditional
NSAIDs, no differences in the incidence of stroke,
cardiovascular death, or myocardial infarction were
observed compared with placebo.59 Further, the inci-
dence of myocardial infarction in the VIGOR trial
(rofecoxib, 0.5%) was essentially identical to the
CLASS trial (celecoxib, 0.5%; ibuprofen, 0.5%; and
diclofenac, 0.3%). Rofecoxib is generally well tolerated
when administered either alone or in combination
with low-dose aspirin and does not alter the antiplate-
let effect of low-dose aspirin in healthy volunteers.60
However the GI safety profiles for rofecoxib/aspirin
combinations have not yet been established.
A significant advantage of rofecoxib relating to its
pharmacological properties is its demonstrated effi-
cacy in the treatment of acute pain.61– 63 Rofecoxib
was shown to be as efficacious and rapidly acting as
naproxen sodium in a double-blind, randomized,
placebo- and active-comparator-controlled, parallel-
group trial in pain after orthopedic surgery.62 Rofe-
coxib (50 mg) was superior to placebo (P ⬍ 0.05) and
similar to naproxen sodium for all single-dose mea-
sures of pain relief. On days 2 through 5, the rofe-
coxib 50 mg group also used 40% less supplemental
narcotic analgesia (P ⫽ 0.005) and reported less pain
on global evaluations (P ⫽ 0.041) compared with the
placebo group. Rofecoxib (50 mg once daily) has also
been shown to effectively treat postoperative dental
pain.61 Recently a direct comparison was made be-
tween a single dose of celecoxib (200 mg), rofecoxib
(50 mg), or placebo immediately after spinal fusion
April 2002 Volume 323 Number 4

surgery, by monitoring the impact of these coxibs on
subsequent ad lib opiate consumption.46 Although
both rofecoxib and celecoxib produce similar analge-
sic and opiate sparing-effects in the first 4 hours
after surgery, rofecoxib demonstrated extended anal-
gesic and opiate-sparing effects lasting throughout the
24-hr study, consistent with that coxib’s known longer
half life. Another potential advantage of coxib use in
the setting of acute and perioperative pain is rapid-
onset control of inflammation. By minimizing trau-
matic and perioperative swelling, operative fields,
postoperative inflammation, and healing might also be
more manageable.
Pharmacologically speaking, 25 mg of rofecoxib
has been shown to produce an approximately 8 to
12% increase in international normalized ratio at
steady state when co-administered with warfarin.64
Similar observations have been reported with virtu-
ally all known NSAIDs, including celecoxib,44,65– 67
as limited displacement of warfarin from binding
sites on albumin is the common mechanism by
which enhanced warfarin effects can occur, even at
pharmacologically efficacious doses. Although this
increase is not likely to be clinically important in
most patients taking either coxib, use of traditional
NSAIDs with warfarin is almost always contraindi-
cated because of their platelet inhibitory properties.
The complete lack of platelet inhibition and effect on
bleeding time renders both rofecoxib and celecoxib
drugs of choice when an NSAID and warfarin must
be used concomitantly. However vigilant monitoring
of international normalized ratio values should be
conducted when therapy with either coxib is initi-
ated, when doses are changed in patients receiving
warfarin, in all hypoalbuminemic states and, in the
case of celecoxib, when drugs that potentially interact
with cytochrome P450 2C9 are used concomitantly.66,67
In summary, rofecoxib is a potent, rapid acting,
single-dose coxib with low–GI-side-effect profiles
and proven efficacy for OA, acute pain, and dysmen-
orrhea.68 Approval for treatment of RA is still pend-
ing but published studies and practical experiences
are most encouraging. Like all NSAIDs, it is impor-
tant that edema and hypertension be monitored,
especially when high doses and/or high-risk patients
are involved. Rofecoxib, like celecoxib, provides no
thromboprophylaxis and should be used cautiously
in high-risk cardiovascular patients.
Parecoxib and Valdecoxib
The next generation of coxibs will probably have
more specific Cox-2 inhibitory properties, as well as
additional pharmacological advantages such as paren-
teral administration. Parecoxib (Pharmacia Corpora-
tion, Peapack, NJ), will probably be the first available
parenteral coxib under development and is the pro-
drug of the oral formulation valdecoxib (Searle, Inc.,
Skokie, IL).69,70 Currently in initial clinical trials for
the management of acute postsurgical pain, parecoxib
THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
McMurray and Hardy
has been compared with ketorolac (Toradol), the only
currently available non-narcotic parental analgesic for
relief of moderate to severe acute pain. Pharmacoki-
netic studies have shown that parecoxib is converted
to valdecoxib within a short time after administration
by intramuscular or intravenous injection. Both pare-
coxib and valdecoxib are sulfonamide-containing med-
ications, and incidence of allergic cross-reactivity has
yet to be determined. These drugs are nearly 5 times
more Cox-2 specific than celecoxib and might therefore
be predictably comparable with rofecoxib in GI tolera-
bility. Not surprisingly, in clinical trials, parecoxib
compared favorably in efficacy with ketorolac but pro-
duced significantly fewer gastric or duodenal ulcers.
Parecoxib has, in early results, demonstrated the pre-
dicted gastroprotective effects of a coxib while perform-
ing as a parenteral non-narcotic analgesic for severe
postsurgical pain and, therefore, probably for other
acutely painful conditions as well. The oral formula-
tion, valdecoxib, might be expected to perform similar-
ly.69,70 Available pharmacological characteristics of
these drugs are listed in Table 1.
Etoricoxib
Etoricoxib (Merck & Co.) is a potent, rapid-acting,
second-generation coxib under development. Etori-
coxib reportedly has the highest known Cox-2 selectiv-
ity ratio,6 being approximately 3 times as Cox-2 spe-
cific as rofecoxib, valdecoxib, and parecoxib, and
nearly 15 times as Cox-2 specific as celecoxib. In pre-
liminary reports, etoricoxib (MK-0663) rapidly and
selectively inhibited Cox-2 in human whole-blood as-
says in vitro, with an IC50 of 1.1 ⫾ 0.1 ␮mol/L for Cox-2
(LPS-induced prostaglandin E2 synthesis), compared
with an IC50 of 116 ⫾ 8 ␮mol/L for COX-1 (serum
thromboxane B2 generation after clotting of the blood).
Using the ratio of IC50 values (Cox-1/Cox-2), the selec-
tivity ratio for the inhibition of Cox-2 by etoricoxib in
the human whole-blood assay was 106, compared with
values of 35, 30, 7.6, 2.4, and 2.0 for rofecoxib, valde-
coxib, celecoxib, etodolac, and meloxicam, respectively.
Etoricoxib did not inhibit platelet or human recombi-
nant Cox-1 under most assay conditions. Etoricoxib
was a potent inhibitor in models of carrageenan-in-
duced paw edema, carrageenan-induced paw hyperal-
gesia, LPS-induced fever, and adjuvant-induced ar-
thritis in rats, without effects on gastrointestinal
permeability up to a dose of 200 mg/kg/d for 10 days.
Etoricoxib reversed LPS-induced fever in animals by
81% within 2 hours of administration (3 mg/kg) and
showed no effect on GI blood loss, in contrast to lower
doses of diclofenac or naproxen. In summary, etori-
coxib is a potent, rapid-acting coxib under develop-
ment that has the highest selectivity for Cox-2 inhibi-
tion and the lowest inhibition of Cox-1 compared with
all known NSAIDs.6 As with the class of coxibs in
general, the effects of such a highly selective agent on
coagulation and cardiovascular/renal function will
need to be studied carefully.
187
Cox-2 Inhibitors
Conclusion
The elucidation of Cox-2 dependent inflammatory
pathways and the subsequent design of specific in-
hibitors of those pathways, the coxibs, have provided
new weaponry for the antirheumatic arsenal. This
has mainly derived from the proven safety advan-
tages of coxibs in lowering gastrointestinal side ef-
fects. Clinical trials have demonstrated the efficacy
of coxibs to be completely comparable with tradi-
tional NSAIDs, and pharmacoeconomics suggest fa-
vorable cost/benefit ratios with these agents com-
pared with traditional NSAIDs. Although several
clinical questions (such as use with low-dose aspirin,
risk of thrombosis, edema, and hypertension) re-
main, such issues may be common to all NSAIDs
and must not overshadow the overwhelming impor-
tance of the coxibs relative to their GI safety advan-
tage. Although newer agents may alter the choice, it
is likely that coxibs will become, if they are not
already, the drugs of first choice in the treatment of
acute pain, chronic pain, and most rheumatic condi-
tions in the 21st century. Undoubtedly, the indica-
tions for their use will expand well beyond the cur-
rent FDA guidelines, and many physicians are
already using these medicines adjunctively in man-
aging headache, fever, temporomandibular joint
syndrome, multiple acute and chronic pain syn-
dromes, various inflammatory processes, and a host
of rheumatic conditions. In addition to broadly ex-
panding our everyday knowledge and experience
with prostaglandin pathways and their inhibition,
the coxibs are also providing us daily insights into
both the treatment and the fundamental pathogen-
esis of disorders such as rheumatoid arthritis, Alz-
heimer disease,71 and malignant transformation.72
It is highly unlikely that they will be entirely re-
placed by anything in the very near future.
Note added in proof: Recent data from available
valdecoxib prescribing information has been added
to Table 1 without alterations to the text.
References
1. Smith WL, DeWitt DL, Garavito RM. Cyclooxygenases:
structural, cellular, and molecular biology. Annu Rev Bio-
chem 2000;69:145– 82.
2. Everts B, Wahrborg P, Hedner T. COX-2-Specific inhibi-
tors–the emergence of a new class of analgesic and anti-
inflammatory drugs. Clin Rheumatol 2000;19:331– 43.
3. Katori M, Majima M. Cyclooxygenase-2: its rich diversity of
roles and possible application of its selective inhibitors. In-
flamm Res 2000;49:367–92.
4. Dannhardt G, Kiefer W. Cyclooxygenase inhibitors - current
status and future prospects. Eur J Med Chem 2001;36:109 –26.
5. Verburg KM, Maziasz TJ, Weiner E, et al. Cox-2-specific
inhibitors: definition of a new therapeutic concept. Am J Ther
2001;8:49 – 64.
6. Riendeau D, Percival MD, Brideau C, et al. Etoricoxib
(MK-0663): preclinical profile and comparison with other
agents that selectively inhibit cyclooxygenase-2. J Pharmacol
Exp Ther 2001;296:558 – 66.
188
7. Van Hecken A, Schwartz JI, Depre M, et al. Comparative
inhibitory activity of rofecoxib, meloxicam, diclofenac, ibupro-
fen, and naproxen on COX-2 versus COX-1 in healthy volun-
teers. J Clin Pharmacol 2000;40:1109 –20.
8. Warner TD, Giuliano F, Vojnovic I, et al. Nonsteroid drug
selectivities for cyclo-oxygenase-1 rather than cyclo-oxygenase-2
are associated with human gastrointestinal toxicity: a full in
vitro analysis. Proc Natl Acad Sci U S A 1999;96:7563–8.
9. Singh G. Recent considerations in NSAID drug gastropathy.
Am J Med 1998;105(1B):319 – 85.
10. Dicpinigaitis PV. Effect of the cyclooxygenase-2 inhibitor
celecoxib on bronchial responsiveness and cough reflex sen-
sitivity in asthmatics. Pulm Pharmacol Ther 2001;14:93–97.
11. Stevenson DD, Simon RA. Lack of cross-reactivity between
rofecoxib and aspirin in aspirin-sensitive patients with
asthma. J Allergy Clin Immunol 2001;108:47–51.
12. Pang L, Pitt A, Petkova D, et al. The Cox-1/Cox-2 balance
in asthma. Clin Exp Allergy 1998;28:1050 – 8.
13. Patterson R, Bello AE, Lefkowith J. Immunologic tolera-
bility profile of celecoxib. Clin Ther 1999;21:2065–78.
14. Carrillo-Jimenez R, Nurnberger M. Celecoxib-induced
pancreatitis and hepatitis. Arch Intern Med 2000;160:553– 4.
15. Galan MV, Gordon SC, Silverman AL. Celecoxib-induced
cholestatic hepatitis. Ann Intern Med 2001;134:254.
16. Silverstein FE, Faich G, Goldstein JL, et al. Gastrointes-
tinal toxicity with celecoxib vs nonsteroidal anti-inflamma-
tory drugs for osteoarthritis and rheumatoid arthritis: the
CLASS study: A randomized controlled trial. Celecoxib Long-
term Arthritis Safety Study. JAMA 2000;284:1247–55.
17. Cribb AE, Lee BL, Trepanier LA, et al. Adverse reactions
to sulphonamide and sulphonamide-trimethoprim antimicro-
bials: clinical syndromes and pathogenesis. Adverse Drug
React Toxicol Rev 1996;15:9 –50.
18. Brater DC, Harris C, Redfern JS, et al. Renal effects of
cox-2-selective inhibitors. Am J Nephrol 2001;21:1–15.
19. Eras J, Perazella MA. NSAIDs and the kidney revisited:
are selective cyclooxygenase-2 inhibitors safe? Am J Med Sci
2001;321:181–90.
20. Whelton A, Fort JG, Puma JA, et al. Cyclooxygenase-2-
specific inhibitors and cardiorenal function: a randomized,
controlled trial of celecoxib and rofecoxib in older hyperten-
sive osteoarthritis patients. Am J Ther 2001;2:85–95.
21. Pfister AK, Crisall RJ, Carter WH. Cox-2 Inhibition and
renal function. Ann Intern Med 2001;134:1077– 8.
22. Graham MB. Acute renal failure related to high-dose cele-
coxib. Ann of Int Med 2001;135:69 –70.
23. Swan SK, Rudy DW, Lasseter KC, et al. Effect of cyclooxy-
genase-2 inhibition on renal function in elderly persons re-
ceiving a low-salt diet. A randomized, controlled trial. Ann
Intern Med 2000;133:1–9.
24. Catella-Lawson F, McAdam B, Morrison BW, et al. Effects
of specific inhibition of cyclooxygenase-2 on sodium balance,
hemodynamics, and vasoactive eicosanoids. J Pharmacol Exp
Ther 1999;289:735– 41.
25. Salvati P, Lamberti E, Ferrario R, et al. Long-term
thromboxane synthesis inhibition prolongs survival in mu-
rine lupus nephritis. Kidney Int 1995;47:1168 –75.
26. Tomasoni S, Noris M, Zappella S, et al. Upregulation of
renal and systemic Cox-2 in patients with active lupus ne-
phritis. J Am Soc Nephrol 1998;9:1202–12.
27. Daniels B, Seidenberg B. Cardiovascular safety profile of rofe-
coxib in controlled clinical trials. Arthritis Rheum 1999;42:435–6.
28. Crofford LJ, Oates JC, McCune WJ, et al. Thrombosis in
patients with connective tissue diseases treated with specific
cyclooxygenase 2 inhibitors. A report of four cases. Arthritis
Rheum 2000;43:1891– 6.
April 2002 Volume 323 Number 4

29. Bombardier C, Laine L, Reicin A, et al. Comparison of
upper gastrointestinal toxicity of rofecoxib and naproxen in
patients with rheumatoid arthritis. VIGOR Study Group.
N Engl J Med 2000;343:1520 – 8.
30. Catella-Lawson F, Crofford LJ. Cyclooxygenase inhibition
and thrombogenicity. Am J Med 2001;110:28 –32.
31. McMurray RW. Cyclooxygenase 2 inhibition and thrombo-
sis: comment on the Article by Crofford et al. Arthritis Rheum
2001;44:984.
32. McAdam FB, Catella-Lawson F, Mardini IA, et al. Sys-
temic biosynthesis of prostacyclin by Cox-2: the human phar-
macology of a selective inhibitor of Cox-2. Proc Natl Acad Sci
1999;96:272–7.
33. Mukherjee D, Nissen SE, Topol EJ. Risk of cardiovascular
events associated with selective Cox-2 inhibitors. JAMA
2001;286:954 –9.
34. Chancellor JV, Hunsche E, de Cruz E, et al. Economic
evaluation of celecoxib, a new cyclo-oxygenase 2 specific in-
hibitor, in Switzerland. Pharmacoeconomics 2001;19:59 –75.
35. Peloso PM, Scheiman JM. The economic implications of cy-
clooxygenase-2-specific inhibitors Am J Med 2001;110:50–54.
36. Zabinski RA, Burke TA, Johnson J, et al. An economic
model for determining the costs and consequences of using
various treatment alternatives for the management of arthri-
tis in Canada. Pharmacoeconomics 2001;19:49 –58.
37. Burke TA, Zabinski RA, Pettitt D, et al. A framework for
evaluating the clinical consequences of initial therapy with
NSAIDs, NSAIDs plus gastroprotective agents, or celecoxib in
the treatment of arthritis. Pharmacoeconomics 2001;19:33– 47.
38. Haglund U, Svarvar P. The Swedish ACCES model: pre-
dicting the health economic impact of celecoxib in patients
with osteoarthritis or rheumatoid arthritis. Rheumatology
(Oxford) 2000;39:51–56.
39. Tive L. Celecoxib clinical profile. Rheumatology (Oxford).
2000;39:21–28; discussion 57–59.
40. Bensen WG, Zhao SZ, Burke TA, et al. Upper gastrointesti-
nal tolerability of celecoxib, a COX-2 specific inhibitor, com-
pared to naproxen and placebo. J Rheumatol 2000;27:1876– 83.
41. Taylor B, van de Wal B. Safety of celecoxib vs other non-
steroidal anti-inflammatory drugs. JAMA 2000;284:3123– 4.
42. McKenna F, Borenstein D, Wendt H, et al. Celecoxib
versus diclofenac in the management of osteoarthritis of the
knee. Scand J Rheumatol 2001;30:11–18.
43. Gottlieb S. FDA refuses companies’ request to drop ulcer
warning. BMJ 2001;322:385.
44. Linder JD, Monkemuller KE, Davis JV, et al. Cyclooxy-
genase-2 inhibitor celecoxib: a possible cause of gastropathy
and hypoprothrombinemia. South Med J 2000;93:930 –2.
45. Mohammed S, Croom DW. Gastropathy due to celecoxib, a
Cox-2 inhibitor. N Engl J Med 1999;340:2005– 6.
46. Gimbel JS, Brugger A, Zhao W, et al. Efficacy and toler-
ability of celecoxib versus hydrocodone/acetaminophen in the
treatment of pain after ambulatory orthopedic surgery in
adults. Clin Ther 2001;23:228 – 41.
47. Reuben SS, Connelly NR. Postoperative analgesic effects
of celecoxib or rofecoxib after spinal fusion surgery. Anesth
Analg 2000;91:1221–5.
48. Witter J. Cox-2 & 2X: improved safety? FDA Arthritis Advisory
Committee, 2001 Feb 7. Available at: URL: http://www.fda.gov/
ohrms/dockets/ac/01/slides/3677s1_02_witter/index.htm.
49. Goldkind L. GI review— highlights of the CLASS study.
FDA Arthritis Advisory Committee, 2001 Feb 7. Available
at: URL: http://www.fda.gov/ohrms/dockets/ac/01/slides/
3677s1_03_ goldkind/index.htm.
50. Lynch PM. COX-2 inhibition in clinical cancer prevention.
Oncology (Huntingt) 2001;15:21–26.
THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
McMurray and Hardy
51. Sjodahl R. Nonsteroidal anti-inflammatory drugs and the
gastrointestinal tract. Extent, mode, and dose dependence of
anticancer effects. Am J Med 2001;110(1A);66S– 69S.
52. Koki AT, Leahy KM, Masferrer JL. Potential utility of
COX-2 inhibitors in chemoprevention and chemotherapy. Ex-
pert Opin Investig Drugs 1999;8:1623–38.
53. Schnitzer FJ, Truitt K, Fleischmann R, et al. The safety
profile, tolerability, and effective dose range of rofecoxib in the
treatment of rheumatoid arthritis. Clin Ther 1999;21:1688–9.
54. Ehrich EW, et al. Characterization of rofecoxib as a Cox-2
isoform inhibitor and demonstration of analgesia in the den-
tal pain model. Clin Pharmacol Ther 1999;65:336 – 47.
55. Hillson JL, Furst DE. Rofecoxib. Expert Opin Pharmaco-
ther 2000;1:1053– 66.
56. Acevedo E, Castaneda O, Ugaz M, et al. Tolerability
profiles of rofecoxib (Vioxx) and Arthrotec. A comparison of
six weeks treatment in patients with osteoarthritis. Scand
J Rheumatol 2001;30:19 –24.
57. Flynn J, Margolis S. The link between arthritis and heart
disease. Johns Hopkins Arthritis White Paper 2001:34 –39.
58. McEntegart A, Capell HA, Creran D, et al. Cardiovascular
risk factors, including thrombotic variables, in a population
with rheumatoid arthritis. Rheumatology 2001;40:640–4.
59. Cannon GW, Caldwell JR, Holt P, et al. Rofecoxib, a specific
inhibitor of Cox-2, with clinical efficacy comparable with that of
diclofenac sodium. Arthritis Rheum 2000;43:978– 87.
60. Greenberg HE, Gottesdiener K, Huntington M, et al. A
new cyclooxygenase-2 inhibitor, rofecoxib (VIOXX), did not
alter the antiplatelet effects of low-dose aspirin in healthy
volunteers. J Clin Pharmacol 2000;40:1509 –15.
61. Malmstrom K, Daniels S, Kotey P, et al. Comparison of
rofecoxib and celecoxib, two Cox-2 inhibitors, in postoperative
dental pain: A randomized, placebo- and active-comparator-
controlled clinical trial. Clin Ther 1999;21:1653– 63.
62. Reicin A, Brown J, Jove M, et al. Efficacy of single-dose
and multidose rofecoxib in the treatment of post-orthopedic
surgery pain. Am J Orthop 2001;30:40 – 8.
63. Morrison BW, Fricke J, Brown J, et al. The optimal
analgesic dose of rofecoxib: Overview of six randomized con-
trolled trials. J Am Dent Assoc 2000;131:1729 –37.
64. Schwartz JI, Bugianesi KJ, Ebel DL, et al. The effect of
rofecoxib on the pharmacodynamics and pharmacokinetics of
warfarin. Clin Pharmacol Ther 2000;68:626 –36.
65. Shorr RI, Ray WA, Daugherty JR, et al. Concurrent use of
NSAIDs and oral anticoagulants places elderly persons at
high risk for hemorrhagic peptic ulcer disease. Arch Intern
Med 1993;153:1665–70.
66. Haase KK, Rojas-Fernandez CH, Lane L, et al. Potential
interaction between celecoxib and warfarin. Ann Pharmaco-
ther 2000;34:666 –7.
67. Mersfelder T, Stewart LR. Warfarin and celecoxib interac-
tion. Ann Pharmacother 2000;34:325–7.
68. Morrison BW, Daniels SE, Kotey P, et al. Rofecoxib, a
specific Cox-2 inhibitor, in primary dysmenorrhea: A ran-
domized controlled trial. Obstet Gynecol 1999;94:504 – 8.
69. Jain KK. Evaluation of intravenous parecoxib for the relief
of acute post-surgical pain. Expert Opin Investig Drugs 2000;
9:2717–23.
70. Talley JJ, Bertenshaw SR, Brown DL, et al. N-[[(5-methyl-
3-phenylisoxazol-4-yl)-phenyl]sulfonyl]propanamide, sodium
salt, parecoxib sodium: A potent and selective inhibitor of COX-2
for parenteral administration. J Med Chem 2000;43:1661–3.
71. Hull M, Lieb K, Fiebich BL. Anti-inflammatory drugs: a
hope for Alzheimer’s disease? Expert Opin Investig Drugs
2000;9:671– 83.
72. Lynch PM. COX-2 inhibition in clinical cancer prevention.
Oncology (Huntingt) 2001;Mar;15(3 Suppl 5):21– 6.
189