A comparison of valdecoxib and naproxen in the treatment of rheumatoid arthritis symptoms

Clinical

Therapeutics/Volume 28, Number 2, 2006

A Comparison of Valdecoxib and Naproxen in the Treatment of Rheumatoid Arthritis Symptoms GaryW. Williams, MD, PhD1; Alan J. Kivitz, MD2; Mark T. Brown, MD3; and Kenneth M. Verburg, PhD 3

1Scripps Clinic, LaJolla, California; 2AItoona Center for Clinical Research, Duncansville, Pennsylvania; and 3Pfizer Global Research and Development, Ann Arbor, Michigan ABSTRACT Objectives: The primary aim of this work was to compare the efficacy of valdecoxib 10, 20, and 40 mg QD with that of placebo and naproxen 500 mg BID in patients with rheumatoid arthritis (RA). The overall safety and tolerability profiles of valdecoxib and naproxen were also compared. Methods: A 12-week, multicenter, randomized, doubleblind, parallel-group, placebo- and active-controlled study was performed in patients with adult-onset RA whose disease was in a flare state after discontinuing NSAIDs or other analgesics. Patients were randomly assigned to valdecoxib 10, 20, or 40 mg QD, naproxen 500 mg BID, or placebo. The primary efficacy measures were the American College of Rheumatology (ACR) 20% responder index (ACR-20), physicians' assessments of tender/painful joint count and swollen joint count, and patients' and physicians' global assessments of disease activity. Adverse events, clinical laboratory data, and vital signs were assessed by the investigator and compared between treatment groups to evaluate overall tolerability and safety. Results: A total of 1093 patients were randomized to receive either valdecoxib 10 mg QD (n = 226), valdecoxib 20 mg QD (n = 219), valdecoxib 40 mg QD (n = 209), naproxen 500 mg BID (n = 219), or placebo (n = 220). At all time points, the proportion of ACR-20 responders was significantly higher in the valdecoxib groups than the placebo group at weeks 2 (10 mg, P < 0.001; 20 mg, P = 0.008; 40 mg, P = 0.004), 6 (all, P < 0.001), and 12 (10 mg, P = 0.006; 20 mg, P = 0.004; 40 mg, P < 0.001). Similarly, at all time points, the proportion of ACR-20 responders was significantly higher in the naproxen 500-mg group than the placebo group (all time points, P < 0.001). In addition, mean changes in the number of tender/ painful joint counts were significantly greater in the valdecoxib groups than the placebo group at weeks 2 204

(all, P < 0.001), 6 (10 rag, P = 0.002; 20 and 40 rag, P < 0.001), and 12 (10 rag, P = 0.004; 20 rag, P = 0.012; 40 mg, P < 0.001). Naproxen treatment was also associated with greater reductions in tender/painful joint count than placebo (all, P < 0.001). Mean changes in swollen joint count decreased at all time points in all groups, with significantly greater changes in the valdecoxib and naproxen treatment groups than the placebo group (valdecoxib 20 and 40 mg: week 6, P = 0.014 and P = 0.003, respectively; naproxen: week 2, P = 0.014; week 6, P = 0.015; week 12, P = 0.030). Physicians' global assessments of disease activity scores were significantly lower in the valdecoxib (10 mg: weeks 2 and 6, P < 0.001; week 12, P = 0.001; 20 and 40 mg: all weeks, P < 0.001) and naproxen (all time points, P < 0.001) treatment groups than the placebo group. Adverse events were reported by 45.5% patients in the placebo group, 51.8% in the valdecoxib 10 mg QD group, 58.0% in the valdecoxib 20 mg QD group, 56.9% in the valdecoxib 40 mg QD group, and 62.6% in the naproxen 500 mg BID treatment group. Conclusions: Valdecoxib 10, 20, and 40 mg QD were efficacious for treating the signs and symptoms of RA in these patients. The efficacy of valdecoxib 20 and 40 mg QD was not significantly different from that of naproxen 500 mg BID. Valdecoxib was generally well tolerated in this stud> (Clin Tber. 2006; 28:204-221) Copyright 9 2006 Excerpta Medica, Inc.

This work was presented in part at the American College of Rheumatology 2002 Annual Scientific Meeting, October 24-29, 2002, New Orleans, Louisiana.

Accepted for publicationJanuary 4, 2006. Express Track online publication January 30, 2006. doi:l 0.1016/j.clinthera.2006.01.016 0149-2918/06/$19.00 Printed in the USA. Reproduction in whole or part: is not: permitted. Copyright 9 2006 Excerpta Medica, Inc.

Volume 28 Number 2

G.W. Williams et al.

Key words: rheumatoid arthritis, valdecoxib, cyclooxygenase-2 (COX-2), naproxen.

INTRODUCTION

Current pharmacologic options for the treatment of joint pain and inflammation of rheumatoid arthritis (RA) include nonselective (conventional) NSAIDs and selective cyclooxygenase-2 (COX-2)inhibitors. Nonselective NSAIDs inhibit the activity of both cyclooxygenase isozymes, COX-1 and COX-2, whereas the selective COX-2 agents inhibit only COX-2 at full therapeutic doses. 1 Inhibition of COX-2 underlies the anti-inflammatory activity of these agents, whereas inhibition of COX-1 by nonselective drugs is associated with an elevated risk for upper gastrointestinal (GI) ulceration and bleeding. 1,2 Patients with RA who are taking conventional NSAIDs are at especially high risk for such complications. 2 Clinical findings support the improved upper GI safety and tolerability profile of the selective COX-2 inhibitor valdecoxib over those of naproxen, diclofenac, and ibuprofen in patients with RA or osteoarthritis (OA). 3-s At doses of 10, 20, and 40 mg daily, valdecoxib has also been shown to provide greater efficacy (as assessed by the American College of Rheumatology [ACR] 20% responder index [ACR-20]) than placebo and similar efficacy when compared with naproxen in >1000 patients with RA. 9 In addition, results from a 26-week study indicated that valdecoxib 20 or 40 mg QD provided sustained analgesic and anti-inflammatory efficacy in >700 RA patients. 1~Furthermore, valdecoxib has been shown to be effective in the treatment of the signs and symptoms of OA. 3,4 Recently, there has been much debate regarding the cardiovascular (CV) safety of selective COX-2 inhibitots and conventional NSAIDs. u-13 A pooled analysis of the CV safety of valdecoxib did not suggest an additional risk of thromboembolic CV events compared with naproxen, ibuprofen, or diclofenac treatment in arthritis patients for up to 24 weeks of treatment. 14 However, 2 studies in patients undergoing coronary artery bypass graft surgery found a higher risk of thromboembolic CV events in patients receiving parecoxib (the parenteral prodrug of valdecoxib) followed by valdecoxib than in those receiving placebo. 13,15 No such increased risk has been observed in a large trial of patients undergoing major noncardiac general surgery.16 In 2005, the US Food and Drug AdFebruary 2006

ministration (FDA) requested that sales of valdecoxib be suspended in the United States. The FDA concluded that, in light of the potential increased CV risk for all selective or nonselective NSAIDs, the additional increased rate of rare, serious skin reactions with valdecoxib warranted suspension of the product. 17 Similarly, sales of valdecoxib were suspended in Europe at the request of the European Medicines Evaluation Agency (EMEA). The present study, completed before the voluntary suspension of sales in the United States and Europe, was primarily undertaken to compare the efficacy of valdecoxib 10, 20, and 40 mg QD with that of placebo and naproxen 500 mg BID in patients with RA. The overall safety and tolerability profiles of valdecoxib and naproxen were also compared. METHODS Patients

Patients were eligible to participate in the study if they were aged ->18 years, had been diagnosed ->6 months previously with adult-onset RA (as defined by the 1987 ACR criterialS), and classified with a functional capacity between I and III. 19 Eligible patients were required to be stable with use of NSAID therapy, and functional capacity must have remained stable for ->1 month before the screening visit. In addition, all patients were required to be in a flare state within 2 to 7 days after discontinuation of NSAIDs, full-dose aspirin, or celecoxib therapy; within 4 to 7 days after discontinuation of oxaprozin and/or piroxicam; or within 4 days of discontinuing rofecoxib for RA. A flare state was defined as a rating of fair, poor, or very poor on both the patients' and physicians' global assessments of disease activity at the baseline visit; ->6 tender/painful joints and an increase of 2 tender/ painful joints (or 20% increase in the number of tender/painful joints, whichever was greater); and ->3 swollen joints, with an increase of ->2 swollen joints (or ->20% increase in the number of swollen joints, whichever was greater) compared with those observed at the screening visit. Patients were also required to experience ->45 minutes of morning stiffness at baseline, with an increase in the duration of morning stiffness of ->15 minutes compared with the screening visit, or a measurement of ->40 mm on the patients' assessment of arthritis pain on a 100-mm visual analog scale (VAS) (by which patients rated their pain between 0 [no pain] and 100 [maximally severe pain[) at 205

Clinical

Therapeutics

baseline, with an increase of ->10 ram, or ->20%, whichever was greater, compared with the screening visit. Patients were excluded from the study if they had any inflammatory arthritis other than RA, or a secondary noninflammatory type of arthritis (such as OA or fibromyalgia) that might interfere with the evaluation of the effect of study medication on RA symptoms. Exclusion criteria also included initiation of or change in dose regimen for gold salts or antimalarials within the past 4 months; sulfasalazine (>3 g/d), azathioprine, penicillamine, methotrexate (except for dosages up to 25 mg/wk), etanercept, leflunomide, combination therapies, or antibiotics used in RA treatment within 12 weeks; glucosamine/chondroitin within 4 weeks before receiving the initial dose of study medication; oral corticosteroids (>10 mg prednisone or equivalent) within 4 weeks of the initial dose of study medication; corticosteroid injection (IM, intra-arterial, or soft tissue) within 8 weeks of the initial dose of study medication; exposure to any antineoplastic agents (except methotrexate up to 25 mg/wk or azathioprine) for RA within 12 weeks of the initial dose of study medication; use of any nonselective NSAID (including full-dose aspirin or celecoxib) within 48 hours, or any analgesic (including acetaminophen) within 24 hours before the baseline visit. Patients who were taking -<325 mg/d of aspirin for reasons other than arthritis treatment (and for ->30 days before the first dose of study medication) were permitted to continue aspirin therapy throughout the study. Patients were also excluded if they met any of the following criteria: diagnosed or treated for esophageal, gastric, pyloric channel, or duodenal ulceration within 30 days before the first dose of study medication; active GI disease, a chronic or acute renal or hepatic disorder, or significant coagulation defect; receipt of warfarin within 30 days before the first dose of study medication; or use of lithium. Any patient with abnormal liver function tests, uncontrolled diabetes or hypertension, or known hypersensitivity to selective COX-2 inhibitors, lactose, or conventional NSAIDs was excluded. Patients were excluded if they were pregnant or breastfeeding.

Study Design This was a 12-week, multicenter, randomized, doubleblind, parallel-group, placebo- and active-controlled study. Patients were drawn from a total of 131 investigators at 225 study sites in North America and 206

South America. There were 5 investigators in Brazil, 2 investigators in Mexico, 9 investigators in Canada, and 115 investigators in the United States. The study protocol was reviewed and approved by the appropriate institutional review boards or ethics committees, and the study was conducted in accordance with ethical principles based on the Declaration of Helsinki. All participating patients provided written informed consent during the pretreatment screening period before undergoing any study procedures. Patients whose RA was in a flare state (as defined previously) after discontinuation of NSAIDs or other analgesics, and who met all inclusion criteria and none of the exclusion criteria, were randomized to receive valdecoxib 10, 20, or 40 mg QD, naproxen 500 mg BID, or placebo for 12 weeks. Patients were assigned to study group in the order in which they were enrolled, according to a computer-generated randomization schedule prepared before the start of the study. All study medications were self-administered orally according to the prescribed regimen. Patients assigned to valdecoxib 10 mg QD took 2 valdecoxib 5-rag tablets QD and 1 placebo capsule BID to match the naproxen capsule. Similarly, patients taking valdecoxib 20 mg and 40 mg took two 10- and 20-rag tablets QD, respectively, plus 1 placebo capsule BID. All valdecoxib tablets were identical in appearance and size. Patients taking naproxen took 2 placebo tablets QD to match the valdecoxib tablets and 1 naproxen capsule BID. Patients randomized to placebo took 2 placebo tablets QD to match the valdecoxib tablets and 1 placebo capsule BID to match the naproxen capsule. Efficacy and safety assessments were performed at screening, baseline, and weeks 2, 6, and 12, or at withdrawal/early termination from the study.

Efficacy Measures The primary assessments of efficacy were the ACR-20 responder index, physicians' and patients' global assessments of disease activity, and physicians' assessments of tender/painful joint count and number of swollen joints. A patient was classified as an ACR-20 responder if the number of both tender and swollen joints decreased by ->20% and there was a reduction of ->20% from the baseline score in ->3 of the following 5 assessments: patients' assessments of arthritis pain (using 100-ram VAS); patients' global assessments of disease activity; Volume 28 Number 2

G.W. Williams et al.

physicians' global assessments of disease activity; patients' assessments of physical function (modified Health Assessment Questionnaire 2~ [mHAQ], in which the ability to perform activities within 8 areas of daily function is graded according to a 4-point scale [0 = without any difficulty; 1 = with some difficulty; 2 = with much difficulty; 3 = unable to perform] and according to whether help from another person or use of an assistive device is required); or the acute-phase reactant, C-reactive protein (CRP). Patients' and physicians' global assessments of disease activity were each scored independently on a 5-point scale (1 = very good; 2 = good; 3 = fair; 4 = poor; 5 = very poor) according to the perception of the global impact of RA symptoms on all aspects of function. Results were calculated as the change from baseline to weeks 2, 6, and 12, with negative values for mean changes representing improvement. The number of tender/painful and swollen joints was based on assessment of a predefined set of 68 (tender/painful) or 66 (swollen) joints in the upper body, upper extremities, and lower extremities. The only difference between the sets of assessed joints was the exclusion of right and left hip joints for swelling. Secondary efficacy assessments included patients' assessments of arthritis pain (using 100-mm VAS); patients' assessments of physical function (using mHAQ); duration of morning stiffness; CRP; and incidence of withdrawal due to treatment failure. Before any other assessments were performed during the visit, patients were required to answer all mHAQ questions. Duration of morning stiffness was defined as the patients' response to a question regarding the time between awakening and reaching maximal flexibility associated with typical activities over 24 hours of observation. The duration was recorded in hours, rounded to the nearest quarter hour. CRP levels were determined from blood samples taken at each clinic visit. Exploratory analyses included the ACR-50 and ACR-70 responder criteria at weeks 2, 6, and 12. The ACR-50 and ACR-70 response criteria are analogous to the ACR-20, with the exception of necessary reductions of ->50% and ->70%, respectively. Safety Assessments General clinical safety was monitored by the incidence of adverse events, including any signs or symptoms, whether related or unrelated to the condition under study; any clinically significant laboratory abFebruary 2006

normality, or any abnormality detected during physical examination that developed or increased in severity during the course of the study. Signs and symptoms were recorded at the baseline visit for each patient. Any changes in preexisting or new signs or symptoms were recorded and graded by the investigator as mild (causing no limitation of usual activities), moderate (causing some limitation of usual activities), or severe (causing inability to perform usual activities). Laboratory evaluations, including standard biochemistry, hematology, and urinalysis assessments, were performed at the screening visit, within 7 days of the baseline visit, and at weeks 2, 6, and 12. A serious adverse event was defined as any event that was fatal or life threatening (in the investigator's opinion), was or caused a persistent disability/incapacity, required prolonged inpatient hospitalization, or may have jeopardized the patient or required medical or surgical intervention to prevent another outcome defined as serious. Statistical Analyses Baseline comparison of treatment groups included all randomized patients. The homogeneity of treatment groups with regard to baseline demographic, vital sign, laboratory, medical history, treatment history, and RA assessments was evaluated using either 2-way analysis of variance for continuous variables or the Pearson X2 test for categoric variables. All statistical analyses of efficacy were conducted on the intent-to-treat (ITT) sample, defined as the group of patients who were randomized and received ->1 dose of study medication. Efficacy results at week 12 were considered primary, although statistical comparisons were also made using results from weeks 2 and 6. Any efficacy measurements missing at weeks 2, 6, or 12 were imputed using a last-observation-carried-forward (LOCF) approach. For ACR-20 response criteria, pairwise comparisons and linear trend test (excluding naproxen) were performed using the Cochran-Mantel-Haenszel test stratified by center. All other primary and secondary efficacy measures (except withdrawal due to treatment failure) were analyzed using least squares (LS) mean changes from baseline. Because all efficacy measures other than ACR-20 response and withdrawal due to treatment failure were based on assessments in which higher scores implied worsening of symptoms, negative values for mean changes represented improvement. 207

Clinical Therapeutics Between-group comparisons were based on linear trend analyses (excluding naproxen) and pairwise comparisons carried out by analysis of covariance, with center and treatment group as factors and baseline as the covariate. Results of the primary pairwise efficacy comparisons (valdecoxib 20 mg QD vs placebo and valdecoxib 40 mg QD vs placebo) were interpreted using the Hochberg step-down procedure. 21 LS mean change ratios for each of the valdecoxib treatment groups versus the naproxen treatment group were calculated with 95% CIs. Analyses of categorical change in the primary and secondary efficacy measures (except ACR-20 response and CRP values) were also conducted using linear trend tests (excluding naproxen), and pairwise comparisons were conducted using the Cochran-MantelHaenszel test adjusted for center effects. All statistical analyses of safety were based on the ITT samples. Adverse events were coded and summarized by treatment group. The incidence of adverse events resulting in withdrawal and serious adverse events were tabulated. The sample size for this study was based on the anticipation that 20% of patients in the placebo group and 35% of patients in the active treatment groups would meet the criteria for an ACR-20 response. A sample size of 200 in each treatment group was determined to be sufficient to detect the difference between placebo and active treatment with a power of 80% and (~ = 0.017. In addition, this sample size was also determined to be sufficient to detect differences between the placebo group and each of the valdecoxib groups with ---80% power with regard to mean changes from baseline that were >0.294 for patients' global assessments of disease activity, >0.285 for physicians' global assessments of disease activity, >3.737 in number of tender/painful joints, and >2.733 in number of swollen joints. RES U LTS Patient Characteristics A total of 1093 patients diagnosed with RA were randomized to receive either valdecoxib 10 mg QD (n = 226), valdecoxib 20 mg QD (n = 219), valdecoxib 40 mg QD (n = 209), naproxen 500 mg BID (n = 219), or placebo (n = 220) for 12 weeks. Four hundred forty-two patients (40.4%) withdrew from the study before completion. The most frequent reasons for withdrawal were treatment failure or adverse events.

208

The overall disposition of patients by treatment group is summarized in Figure 1. With regard to demographic characteristics, the groups were well matched with the exception of age (Table I). Mean (SD) age was somewhat higher in the placebo group (58.1 [12.76] years) and lower in the naproxen 500-mg BID treatment group (54.5 [13.07] years) compared with the valdecoxib groups (10 mg, 56.8 [11.61] years; 20 mg, 55.1 [11.79] years; 40 mg, 56.9 [12.94] years; P = 0.015). The groups were well matched with regard to all other baseline characteristics. Efficacy The percentage of patients who achieved an ACR-20 response in the valdecoxib 10-, 20-, and 40-mg QD treatment groups was comparable across all assessments (P = NS), and was significantly greater than the placebo group at weeks 2 (10 mg, P < 0.001; 20 mg, P = 0.008; 40 mg, P = 0.004), 6 (all, P < 0.001), and 12 (10 mg, P = 0.006; 20 mg, P = 0.004; 40 mg, P < 0.001) (Figure 2). The percentages of ACR-20 responders in the valdecoxib 10-, 20-, and 40-mg QD treatment groups were comparable with that of the naproxen 500-mg BID treatment group at weeks 6 and 12 (P = NS). At week 2, there was a significantly greater percentage of ACR-20 responders in the naproxen 500-mg BID treatment group than in any of the valdecoxib treatment groups (10 mg, P = 0.025; 20 mg, P = 0.002; 40 mg, P = 0.005). The percentages of ACR-20 responders were higher in the naproxen group than the placebo group at all time points (all, P < 0.001). The percentage of ACR-50 responders was also significantly greater for valdecoxib 20 mg (12% and 17%) and 40 mg (11% and 16%) than for placebo (3% and 8%) at weeks 2 (both, P < 0.001) and 6 (both, P = 0.018), respectively, but there were no significant treatment differences observed at week 12. In the valdecoxib 40-mg QD group, ACR-50 responder rates at weeks 2, 6, and 12, respectively (11%, 16% and 17%) were not significantly different from those in the naproxen 500-mg BID treatment group (18%, 21%, and 25%). However, ACR-50 responder rates for valdecoxib 10 mg (11% and 14%) were significantly lower than for naproxen 500 mg (18% and 21%) at weeks 2 (P = 0.021) and 6 (P = 0.020), respectively. There was no difference in ACR-70 responders between the valdecoxib 10-mg QD and placebo groups, but there were statistically significant differences beVolume 28 Number 2

G . W . W i l l i a m s et al.

8s.~

# ~ ~~ Z

es m

~

m

E

8

n

~ ~ =

~ S o~ Zo~ x o

m

E

es z

8

= ~ ~s.~

n ([7

~

E~ o J3

~)

cq ii

~ Z ~ m

E

m

o_

8

n ([7

8s.~

~-~ ~

cq cq

0

0

# ~ ~~ Z

~

m

E

m

8 cO

n ([7

0

~ ~ s ._~ ~~

E~~

~ ~-,~ 8 ~ cq 13_ Z

~ r~

m

m

E

,e-

8 [J_

February 2006

209

Clinical T h e r a p e u t i c s

cO -i

o

&

fl

s d

d

cO O

oo

,,

o

~

=..%

0~

00

o,I Oh 0

0 ',.O

Oh Oh '~

o') ',.O 0..I

o') O0 O0

d

o

d

d

o

0

o

~

~

~

~

0

0

~

0

~

~

~

~

0

~

~

~

0

~

0

0

~

0

0

0

0

~

r,.,

= -,,~

~8

=..% ,~ ~

~

0

0

~

~

0

v v

v

v

v

v

v

.-.

S"

"-" o,

0,1

Oh

~

~

v

v

v

v

o

5

00

,,-,,

R-" ~--" "-" o v

v

v ~ ~

-~

0

0

0

~

t-.

E "

o

~

o

~

~

~

~

~

"r"

U

E

,._

~ ,

E

v._.

v ~

:~U

0

~

~

0

~

v

v

v

v

0

~

r-.

C C

E e-

O

-"-- o~ o~ v

E

v

"~"

9

v

d

d r-

d ~i

~

~

0 cd u

II

~5

S

~

E

a~

"~

c.~_

r7

d G "r-

>" ~

O

O

~,

0 >.

0 >.

0 >.

"'~

o

o

o

~

C3 ..2"

210

9N c_ u

~

~~2

E

u

<

.i c

E ~

~8

.~

"-'~ ---~-~ ~ ~ o~._~

m ~ < O

~

U

0

U

II i

m

~

I

I

I

o..

L3

en

Volume 28 Number

2

G.W. Williams et al.

9 [] 9 [] []

601

Valdecoxib 10 Valdecoxib 20 Valdecoxib 40 Naproxen 500 Placebo

mgQD mgQD mg QD mg BID

50

v c c

o

c o3 (3..

Week 2

Week 6

Week 12

Figure 2. American College of Rheumatology 20% responder index at weeks 2, 6, and 12, according to study group. *P < 0.001 versus placebo; tp = 0.008 versus placebo; ~P = 0.004 versus placebo; w = 0.025 versus valdecoxib 10 mg; lip = 0.002 versus valdecoxib 20 mg; gP = 0.005 versus valdecoxib 40 mg; #P = 0 . 0 0 6 versus placebo.

tween the placebo group and the valdecoxib 20- and 40-mg QD groups and the naproxen group at week 12 (P = 0.034, P = 0.010, and P = 0.018, respectively). The only significant difference in ACR-70 responders between the naproxen and valdecoxib groups was in the valdecoxib 40-mg QD group at week 6, in favor of naproxen treatment (P = 0.049). The mean scores for both physicians' and patients' global assessments of disease activity improved in all treatment groups (Table II). For the physicians' assessments, valdecoxib 10, 20, and 40 mg QD were all associated with significantly greater improvements than placebo at weeks 2, 6, and 12, respectively (10 mg: weeks 2 and 6, P < 0.001; week 12, P = 0.001; 20 and 40 mg: all, P < 0.001), as was naproxen 500 mg BID (all time points, P < 0.001). For the patients' assessments as well, all valdecoxib groups were associated with greater improvements than placebo at weeks 2, 6, and 12 (all, P < 0.001), as was naproxen 500 mg BID (all time February 2006

points, P < 0.001). In addition, the naproxen 500-mg BID treatment group experienced significantly greater improvements than the valdecoxib 10-mg QD treatment group at all visits for both the patients' (week 2, P = 0.025; week 6, P = 0.007; week 12, P = 0.021) and physicians' (week 2, P = 0.043; week 6, P = 0.010; week 12, P = 0.047) global assessments of disease activity. Significant reductions in the number of tender/ painful joints were also observed for all active treatments (Table II). The number of tender/painful joints was significantly reduced with valdecoxib treatment compared with placebo at weeks 2 (all, P < 0.001), 6 (10 mg, P = 0.002; 20 and 40 mg, P < 0.001), and 12 (10 mg, P = 0.004; 20 mg, P = 0.012; 40 mg, P < 0.001). In general, the reductions in the number of tender/painful joints were similar across all valdecoxib treatment groups (P = NS), except at weeks 2 and 6, when there was a significantly greater reduction in the valdecoxib 40-mg QD treatment group than the 211

Clinical Therapeutics

e-

0

~

0

0

0

0 0

0

0

0 V

0 V

0 V

0 V

0 V

0 V

c c 0 v

~

u'3

u'3

u'3

0

0

I

I

~

u'3

u'3

~5 I

I

0

I

I

D.0

-i

E C3

E

E

O~

O~

O~

d o.

d o.

V

V

oO v d

E O~

E

E O~

O~

-i

r~

.~

O

c'N

E 0 0

-i

<:

0

0

0 0

u") CN

V

0

d o.

0

0

d o. c ~ O v c5 V 0

Z

e-

0

r

0 t~ t~

I

I

0

0

0

0

0

O

0

0

0

0 0

0

O

~5

~5

~5

~5

c5

c~

V

V

V

V

V

V

r~

,e-e-

E 0

c~ I

I

I

0

0

0

0

0

O

0

0

0

0 0

0

O

~5

~5

~5

~5

c5

V

V

V

V

V

r~

0

E

X

e-

E 0 o,I

v .m .m

e-

e-

I

I

I

0

0

0

0

0

O

0

0

0

0 0

0

O

0 V

0 V

0

0 V

0 V

O v

~5

~5

c5

c~

I

I

r~

et~ e-

E E E -i

I

V

O

0

>

c~

0 -i

I

e-

I

I

I

et~ O_ t'N ,e-....2"

o,I

212

0 w

C C'N

m

= ~'~m~

m

r

.Z3 0

>

~ ~

9, ~ ~

m

.,~ L)

~

~

Volume 28 Number 2

G . W . Williams et al.

o~

GA

O O

0 0

0

O V

0 V

0 V

0

0

"o

I

I

rcO L~ v

',.00040 I

b,O

E E 0

n

b0 E

0

E E

0

.nO --'-J ....j

....j

0 0

0 0

cO 0

o,I o,I

V

V

0

0

XXX

q> O O

0o,lo'3

V

V

>

d oO.O.

<(

Z

0

0

d oO.O.

0 0

0 ~

dOO vdd 0

0 I

I

O rn

0

0

b~

O O

0 0

E

d

d

d

V

V

V

O

0

I

0

~

~

0

~

r-n

S 0

q~ "m

o') I

0 I

I

O rn

0

0

O O

0 0

0

d

d

d

r-n

q.

b~

E

V

O o,I

V

q~ "m rg

>

n (:7

~

q.

O

n (:7

'q" o-,I

q~ m O

n (:7

0

0 ~

>

O O

X

E E

o') O

E

~ d d

0

0

I

I

I

O rn

0

0

O O

0 0

o4 0 0

O v

0 V

q-

b~

V

r-n

0

O q~ "m

o')

"o

0

0 I

I

I

-i e-

.m

e0 O4

v ....2" m

r February 2 0 0 6

m

q~ th rg

>

~-E

q.) q.)

aFFF

i

u

oq

E

213

Clinical Therapeutics

O O >

o~

o~

S O

0

0 0

0

O V

O V

0

0

0

I

I

O

O V

TM

S

d~5

a_>

O4

~

~ 0

~

O ~

~

E b,O

E

EE

E

EE

. ~ 0 0 . ~ 0

C3

O~

0

r-n . ~

E

m > m

>

0 0

0 0

~

0 0

0

0 m

0 ~ m m

<(

V

V

O

m > m m

~ 0 ~ 0 ~ 0 ~

0 0

o O O d O

O

V

~

o

~ ~

0

.-c

0

. . . . 0 ~

m > m

P. m m

E

~ 0 ~ 0 ~

o ~ 1 7 6d

O

E

0

0 ~

0 0

-~

E

9~ 0 0 . ~ 0

~ >

E

V O 0

V O

0

0

6 '~

..O

~ 1 7 6 1 7 6

o

.__~s

O 0

Z

"o

z

~

E

.m --

Oh I

d

o

>m

~

a_< >

m >

E

2

o ~ V

O

O

~

V O ~

V

K

~

O

O

b

O

O

d

d

8

E~

o~ o,I

0

I

0

r-n

O_

O

> II

C3

X

>.~

m

S o

0 0 0 v

E

O V

S

S

O

O

L~

d

0

O o,I

~

:-~E

o

~ _ ~

,,

.

'~"

G_'-

H

O,I ~ I

O

II

O

O

C3

E

O O

O O

d

d

d

"~

--

z ....

d-~ ~ _o

O O

-C

o

o~-O

~,,

._ 2~o

h~b

~.

0 ~

Z~

o "-

V

O

~<~-d

I

-i e-

E

tO

~.~

.m

~ O

v

C)_ ~ .

~

~

0

~,

,~

~

q~

~

~ ~ •

C~

9"o

z

._o-.~

z

~ " ~~ ,~

0

~~

~ ~

,

z

.~

0

....2" m

..O

r

214

9r"

>

ff'SE

~

~

E

Volume 28 N u m b e r 2

G . W . W i l l i a m s et al.

valdecoxib 10-rag QD treatment group (P = 0.011 and P = 0.008, respectively). The decrease in the tender/ painful joint count in the valdecoxib 40-rag QD and naproxen 500-rag BID treatment groups was comparable at weeks 2, 6, and 12 (P = NS). Naproxen treatment was associated with a significantly greater reduction of tender/painful joints than valdecoxib 10 mg QD at weeks 2 and 6 (P = 0.010 and P = 0.006, respectively), but not week 12, and compared with valdecoxib 20 mg QD at weeks 6 and 12 (P = 0.042 and P = 0.028, respectively), but not week 2. The number of swollen joints also decreased from baseline in all groups (Table II). Statistically significant reductions in the valdecoxib 20- and 40-rag QD treatment groups relative to placebo were evident at week 6 only (P = 0.014 and P = 0.003, respectively). Reductions in the number of swollen joints were significantly greater in the naproxen 500-rag BID treatment group than the placebo group at all time points (weeks 2 and 6, P < 0.001; week 12, P = 0.001). The reductions in the naproxen 500-rag BID treatment group were not significantly different from those in the valdecoxib 40-rag QD treatment group at all visits (P = NS) but were significantly greater than in the valdecoxib 10-rag QD treatment group at all visits (week 2, P = 0.014; week 6, P = 0.015; week 12, P = 0.030) and in the valdecoxib 20-rag QD treatment group at week 2 (P = 0.045), but not weeks 6 and 12. All doses of valdecoxib provided significantly greater pain relief than placebo, as indicated by significantly greater reductions in LS mean change scores for patients' assessments of arthritis pain at all time points. Comparisons between valdecoxib and placebo were statistically significant for all doses at all time points (all, P < 0.01). Reductions in VAS scores were significantly greater in the naproxen 500-rag BID treatment group than the valdecoxib 10-rag QD treatment group at weeks 2 and 6 (P = 0.002 and P < 0.001, respectively), the valdecoxib 20-rag QD treatment group at weeks 6 and 12 (P = 0.002 and P = 0.017, respectively), and the valdecoxib 40-rag QD treatment group at week 2 (P = 0.015). Improvements in physical function associated with valdecoxib treatments were suggested by changes in mHAQ scores (Table II) and the duration of morning stiffness. Reductions in mHAQ score were significantly greater with valdecoxib than placebo at all time points and with all doses (10 mg at week 12, P = 0.002; all other comparisons, P < 0.001). Naproxen

February 2006

was also associated with greater reductions in mHAQ score than placebo (all time points, P < 0.001). There were significantly greater changes in the duration of morning stiffness in all valdecoxib groups than placebo at weeks 2 (all, P < 0.001), 6 (all, P < 0.001), and 12 (10 rag, P = 0.003; 20 and 40 rag, P < 0.001). Improvements in morning stiffness were also greater with naproxen than placebo (all time points, P < 0.001). There were no significant between-group differences with regard to LS mean change from baseline in CRP. Mean CRP values for all treatment groups were substantially greater than normal (---8000 pg/L), both at baseline (range, 14,617-21,648 pg/L) and throughout the duration of the study (range, 13,057-25,180 pg/L). The incidence of withdrawal due to treatment failure was significantly higher in the placebo group (92 [41.8%]) than in the valdecoxib 10-rag QD (61 [27.0%]), valdecoxib 20-rag QD (56 [25.6%]), valdecoxib 40-rag QD (48 [23.0%]), and naproxen 500-rag BID (43 [19.6%]) treatment groups (all comparisons, P - 0.001). The incidence of withdrawal due to treatment failure was comparable among all valdecoxib treatment groups (P = NS) and between the valdecoxib treatment groups and the naproxen 500-rag treatment group (P = NS).

Safety Assessments Of the 1093 patients in the ITT sample, a total of 600 (54.9%) reported ---1 adverse event (Table III). Most reported adverse events were mild (278; 46.3%) or moderate (261; 43.5%) in severity. The total incidence of adverse events was significantly greater in the valdecoxib 20-rag QD (58.0%), valdecoxib 40-rag QD (56.9%), and naproxen 500-rag BID (62.6%) treatment groups than in the placebo group (45.5%; P 0.05). The incidence of adverse events was significantly higher in the naproxen 500-rag BID treatment group (62.6%) than in the valdecoxib 10-rag QD treatment group (51.8%; P - 0.05). Adverse events led to the withdrawal of 65 (5.9%) patients, with a significantly greater number of withdrawals due to adverse events occurring in the naproxen 500-rag BID treatment group than in the valdecoxib 10-rag QD and placebo groups (P - 0.05). GI adverse events were reported by 44 (20.0%) patients in the placebo group, 48 (21.2%) in the valdecoxib 10-rag, 56 (25.6%) in the 20-rag, and 57 (27.3%) in the valdecoxib 40-rag QD treatment groups, and by 72 (32.9%) in the naproxen 500-rag BID treatment 215

Clinical Therapeutics

v o

o-1

i

E ii~

m

eI

__

~

m

0

vc

t:~

~

0

v

v

v

v

v

C~ v

v

CO

0

,

u3

~

o5 v 0

O

r<

~

v kO

v

v

v

v

v

v

v

o5

C~ CN

Z

O_

E

"~

~

,

r<

~

.--

0

v

II

v

v

v

v

v

v

v C~

u5

eI

0

C3 E ~-. 0.,I

~

0.,I

* C

CO u'3

eI

"--

~

C3

t'~

O

C

O

~

O

~ L~ v

0

v u5 oq

C~

0

fll

0

o-1

~6

0

v

v

u5

0

o

ei

d

t-

E

O ~J

E

c.i

O O If)

0

>

E

o t-

o

o

b

o

e.-

1~.

o r-

~

=

~

--

0

m

c

0

0

~'~

b~

~ 0

u

216

E o

._

No_ N

~c~z<~n-u ~

oc5 Vl

Vl

o

r~

Volume 28 N u m b e r 2

G.W. Williams et al.

group. The incidence of GI adverse events was significantly higher in the naproxen treatment group versus the placebo and valdecoxib 10-mg QD groups (P 0.05). The most frequently reported GI adverse events were dyspepsia (with no significant between-group differences), abdominal pain, and nausea. The valdecoxib 40-mg QD treatment group experienced a significantly higher incidence of abdominal pain than the placebo group (P - 0.05); and the naproxen 500-mg BID treatment group experienced a significantly higher incidence of nausea than the valdecoxib 10- and 40-mg QD and placebo groups (all comparisons, P 0.05). One patient with a history of colon polyps, bleeding peptic ulcer, and hypertension in the naproxen 500-mg BID treatment group was diagnosed with gastric ulcer and a GI bleed after becoming ill on day 49. Study medication was discontinued on day 51 and the patient was withdrawn from the study. Another patient with a history of hypertension and hysterectomy receiving valdecoxib 40 mg QD was also diagnosed with a GI bleed. Serious adverse events were reported in 20 patients: 5 (2.2%) patients in the valdecoxib 10-mg QD treatment group experienced 11 serious adverse events, 1 (0.5 %) patient in the valdecoxib 20-mg QD treatment group experienced 2 serious adverse events, 3 (1.4%) patients in the valdecoxib 40-mg QD treatment group experienced 5 serious adverse events, 5 (2.3%) patients in the naproxen 500-mg BID treatment group experienced 7 serious adverse events, and 6 (2.7%) patients in the placebo group experienced 7 serious adverse events. Serious GI system disorders were reported in 4 patients (1 patient each in the valdecoxib 10- and 40-mg QD, naproxen 500-mg BID, and placebo groups). Serious myocardial, endocardial, or pericardial and valve disorders were reported in 2 patients in the naproxen treatment group and in 1 patient in the placebo group. Two patients in both the valdecoxib 10 and 20 mg QD treatment groups reported serious central and peripheral nervous system disorders. Serious respiratory system disorders were reported by 2 patients in both the valdecoxib 10-mg QD and placebo groups. One patient in the valdecoxib 20-mg QD treatment group reported a serious CV disorder (aneurysm). General disorders, including back pain, occurred in 2 patients in the valdecoxib 10-mg QD and placebo groups. Chest pain was reported by I patient in the valdecoxib 40-mg QD treatment group and in 1 patient in the placebo group. Two patients in the valdecoxib 40-mg QD group reFebruary 2006

ported a serious cerebrovascular disorder. There were no serious adverse skin reactions. Minor adverse skin reactions were observed in all groups except the valdecoxib 40-mg QD treatment group, and there were no statistically significant differences between any 2 treatment groups. One death, attributed to pulmonary carcinoma, was reported during the study in the valdecoxib 10-mg QD treatment group. The patient was a 56-year-old woman with a 40-pack-year history of smoking and additional diagnoses of pericarditis, presbyopia, hyperlipidemia, psoriasis, hypothyroidism, and anemia. The investigator concluded that the event was not associated with the study medication. The incidence of peripheral edema was significantly higher in the valdecoxib 20- and 40-mg QD treatment groups than in the placebo group (2.3% [5 patients], 2.4% [5 patients], and 0%, respectively; P - 0.05). Peripheral edema was reported by 3 (1.3%) patients in the valdecoxib 10-mg QD treatment group and by 1 (0.5%) patient in the naproxen 500-mg BID treatment group; these incidences were not significantly different from that reported in the placebo group. Significantly greater increases in mean systolic blood pressure (SBP) and mean diastolic blood pressure (DBP) were observed at the final visit in both the valdecoxib 20- and 40-mg QD treatment groups compared with the placebo group (P - 0.05). Significant increases in mean SBP and DBP from baseline to final visit were observed in the valdecoxib 40-mg treatment group compared with the naproxen 500-mg BID treatment group (P - 0.05). Baseline mean SBP values and change from baseline at the final visit, respectively, were 129.6 and -3.0 mm Hg for placebo; 128.8 and -1.0 mm Hg for valdecoxib 10 mg QD; 129.4 and 0.3 mm Hg for valdecoxib 20 mg QD; 130.3 and 0.6 mm Hg for valdecoxib 40 mg QD; and 129.2 and -1.9 mm Hg for naproxen 500 mg BID. Baseline mean DBP values and change from baseline at the final visit were 77.7 and -1.5 mm Hg for placebo; 78.1 and -0.5 mm Hg for valdecoxib 10 mg QD; 78.7 and 0.0 mm Hg for valdecoxib 20 mg QD; 78.5 and 0.6 mm Hg for valdecoxib 40 mg QD; and 78.3 and -1.2 mm Hg for naproxen 500 mg BID. DISCUSSION

The results of this study indicate that valdecoxib, at doses of 10, 20, and 40 mg QD, provided significant efficacy for the treatment of the signs and symptoms of RA in the patients who participated. At the end of 21 7

Clinical Therapeutics

the 12-week study, all 3 doses of valdecoxib demonstrated significantly greater improvement than placebo with regard to primary and secondary efficacy measures, with the exception of the number of swollen joints and CRP levels. There were significant differences in efficacy measures between the valdecoxib 20and 40-mg QD, naproxen 500-mg BID, and placebo groups. The results of this study confirm those published in previous studies of valdecoxib in patients with RA. 9,1~ A 12-week study involving 1099 RA patients produced a nearly identical pattern of results, with valdecoxib 10, 20, and 40 mg QD treatment associated with improvements in disease activity, inflammation, pain, and daily function comparable to those observed with naproxen 500-mg BID treatment. 9 Similarly, resuits from a 26-week study indicated that valdecoxib 20 or 40 mg QD provided sustained analgesic and anti-inflammatory efficacy in RA patients. 1~ Furthermore, the efficacy of valdecoxib 20 and 40 mg QD was comparable with that of diclofenac 75 mg BID as measured by the patients' assessments of arthritis pain (using the VAS) and mHAQ. However, in both the current study and the previous 12-week study, the incidence of GI adverse events was higher in the naproxen 500-mg BID treatment group than in the valdecoxib 10- and 20-mg QD treatment groups. 9 Furthermore, in the previous 26-week study, patients receiving valdecoxib 20 or 40 mg experienced a significantly lower incidence of endoscopically confirmed gastroduodenal ulcers than those receiving diclofenac; moreover, the GI tolerability profile of valdecoxib 20 mg was considered to be more favorable than that of diclofenac. 1~ In a pooled analysis of data from blinded, randomized controlled trials, and longterm, open-label trials, 6 valdecoxib was associated with a significantly lower ulcer complication rate than nonselective NSAIDs (0.68% vs 1.96%; P < 0.05). In addition, results of a pooled analysis found a decrease in dyspepsia and an improvement in upper GI tolerability among patients with RA or OA taking valdecoxib, even at supratherapeutic doses, compared with those taking nonselective NSAIDs over 12 weeks. 7 All doses of valdecoxib were found to be generally well tolerated in the current study. The incidence of adverse events was significantly greater in the valdecoxib 20 and 40 mg QD treatment groups than in the placebo group (P - 0.05). Furthermore, the incidence of adverse events was significantly lower in the valde218

coxib 10-mg QD treatment group than the naproxen 500-mg BID treatment group (P - 0.05). The incidence of peripheral edema was significantly greater in the valdecoxib 20- and 40-mg QD treatment groups than in the placebo group (P - 0.05). The incidence of peripheral edema in the valdecoxib 10-mg QD treatment group was comparable with those of the naproxen and placebo groups. These findings are consistent with a meta-analysis in which valdecoxib (2.2% to 3.4%) was associated with a rate of peripheral edema comparable with that of conventional NSAIDs (3.0%).22 The incidence of peripheral edema associated with valdecoxib in the present study was also comparable with that previously reported for the selective COX-2 inhibitor celecoxib and other NSAIDs. 23,24 In the present study, valdecoxib 20 and 40 mg QD were associated with significantly greater increases in mean SBP and DBP than placebo (P - 0.05). Examination of the data shows that the actual mean blood pressure increases were ---0.6 mm Hg for both valdecoxib doses, and there was no increase in mean DBP for valdecoxib 20 mg QD. The statistical significance of these findings may be explained in part by reductions in mean SBP and DBP of 3.0 and 1.5 mm Hg, respectively, in the placebo group. Furthermore, valdecoxib 40 mg QD was associated with a significantly greater increase in SBP and DBP compared with naproxen (P - 0.05). However, these differences were not considered to be clinically significant. No significant changes in either SBP or DBP were observed in the naproxen treatment group compared with the placebo group. Two cerebrovascular thromboembolic adverse events were reported in this 12-week study, both of which occurred in the valdecoxib 40-mg group. However, a pooled analysis of valdecoxib data in arthritis patients receiving treatment for up to 24 weeks did not suggest an additional risk of CV thromboembolic events compared with naproxen, ibuprofen, or diclofenac treatment. 25 Further studies are required to confirm the long-term safety of selective COX-2 inhibitors. The FDA recently reviewed data from the long-term controlled trials in which the CV risk profiles of both nonselective NSAIDs and selective COX-2 inhibitors were compared. The FDA concluded that although the approved selective COX-2 inhibitors were associated with an increased risk of serious adverse CV events compared with placebo, the controlled clinical trials did not clearly demonstrate that the selective COX-2 inhibitors conferred a greater risk of serious Volume 28 Number 2

G.W. Williams et al.

adverse CV events than nonselective NSAIDs. As a consequence, professional labeling for all prescription NSAIDs was revised to include a boxed warning to highlight the potential increased risk of serious adverse CV events with both nonselective NSAIDs and selective COX-2 inhibitors. At the same time, the FDA requested that sales of valdecoxib be voluntarily suspended in the United States, based on its determination of an unfavorable risk-versus-benefit profile for valdecoxib, given the lack of long-term CV safety data, reports of an increase in rare, serious, potentially lifethreatening skin reactions compared with reported rates of such reactions with the use of other selective COX-2 inhibitors or nonselective NSAIDs, and the lack of demonstrated advantage compared with other nonselective NSAIDs. 17 Furthermore, sales of valdecoxib were suspended in Europe at the request of the EMEA. As mentioned previously, no serious skin reactions were reported in our study. There are some potential limitations to this study. The high placebo response rate may be due to the eligibility requirements of the study, which required patients to discontinue regular analgesic medication before baseline assessment and have RA in a flare state as a criterion for entry. Furthermore, there was a higher dropout rate in the placebo group, with the possibility of initial improvement carried forward in an LOCF design. Finally, with respect to efficacy and safety, the study was too small to define precisely the differences between active treatment groups. Likewise, due to the small number of patients enrolled, no conclusions could be reached comparing infrequent adverse events, including CV thromboembolic events, in the active treatment groups. CONCLUSIONS

Valdecoxib 10, 20, and 40 mg QD were efficacious for treating the signs and symptoms of RA in these patients. The efficacy of valdecoxib 20 and 40 mg QD was not significantly different from that of naproxen 500 mg BID. Valdecoxib was generally well tolerated in this study. A C K N O W L E D G M ENTS

This study was sponsored by Pharmacia Corporation and Pfizer Inc. (New York, New York). The authors wish to acknowledge Juliet Fawcett, PhD, and Miriam Gordon, PhD, for their editorial assistance. February 2006

The following investigators participated in this study: Raymond A. Adelizzi, DO; James D. Anderson, MD; Charles R. Arkin, MD; Karamali A. Bandealy, MD; Martha L. Barnett, MD; Garry E. Bayliss, MD; Ralph E. Bennett, MD; Charles A. Birbara, MD; Joel A. Block, MD; Robert A. Bonebrake, MD; Arthur Bookman, MD; Kristine K. Bordenave, MD; David Borenstein, MD; Gary R. Botstein, MD; A. McKay Brabham III, MD; Richard D. Brasington, MD; James E Brodeur, MD; Alan L. Brodsky, MD; Michael C. Burnette, MD; David Burns, MD; Jeffrey B. Butler, MD; Jaques R. Caldwell, MD; Jennifer J. Capezio, MD; Alfred Cividino, MD; Daniel H. Cohen, MD; Selwyn A. Cohen, MD; Ignacia De La Torre, MD; Fernando De Souza Cavalcanti, MD; Isam A. Diab, MD; Frederick Dietz, MD; Emma Dilorio, MD; Thomas R. Dykman, MD; William M. Edwards, MD; James A. Engelbrecht, MD; Mark P. Ettinger, MD; John J. Fahey, MD; Justus E Fiechtner, MD; Joao Perchiavalli Filho, MD; Roy M. Fleischman, MD; Norman B. Gaylis, MD; Bernard E Germain, MD; Carolyn S. Gleason, MD; Oscar S. Gluck, MD; Leslie Goodman, MD; Richard D. Gordon, MD; Warren E. Greth, MD; James T. Halla, MD; E. Robert Harris, MD; Lawrence E. Hart, MD, MSc; Gerald Ho, MD; Peter A. Holt, MD; Shannon Howe, MD; Kent A. Huston, MD; Aqil P. Iman, MD; James T. Jakes, MD; Cameron B. Jones, MD; Jeffrey L. Kaine, MD; Brian R. Kaye, MD; Michael I. Keller, MD; Alastair C. Kennedy, MD; Jay Kim, MD; Robert A. Kimelheim, DO; Alan J. Kivitz, MD; Steven J. Klein, MD; Karen S. Kolba, MD; Joel M. Kremer, MD; Mark W. Layton, MD; David H. Lehman, MD; Robert Levin, MD; Mauro Leyba, MD; Bruce D. Long, MD; Mitchell B. Lowenstein, MD; Steven J. Maestrello, MD; Raymond L. Malamet, MD; David R. Mandel, MD; Gesabel C. Marques, MD; Rajendra K. Marwah, MD; Myron E Mass, MD; Stephen D. Matthews, MD; Angela McCain, MD; Robert J.R. McKendry, MD; George E. McLaughlin, MD; James I. McMillan, MD; Carlos A Mendoza, MD; Christopher R. Morris, MD; Carter V. Multz, MD; Ann Myers, MD; Howard L. Offenberg, MD; Brian Peck, MD; Geraldo Pinheiro, MD; Janet Pope, MD; Akavaram N. Reddy, MD; Richard W. Reese, MD; Jude E Rodrigues, MD; Adam M. Rosen, MD; Stephen C. Ross, MD; Sanford H. Ruth, MD; Bruce M. Rothschild, MD; Eric M. Ruderman, MD; Anthony S. Russell, MD; Joel E. Rutstein, MD; Constantine K. Saadeh, MD; Marshall R. Sack, MD; 219

Clinical Therapeutics

Philippe A. Saxe, MD; Jan I. Schulz, MD; Timothy J. Schwartz, MD; John W. Scott, MD; Antonio Scotton, MD; Craig D. Scoville, MD; Anthony Serba, MD; Leonard H. Serebro, MD; Sandra L. Sessoms, MD; Yvonne Sherber, MD; William J. Shergy, MD; David H. Sikes, MD; Suthin Songcharoen, MD; Maria C. Sosenko, MD; Eugene S. Spiotta, Jr., MD; C. Ruffin Stephenson, M D ; J. Steven Strong, M D ; Peter Szachnowski, MD; Nehemiah T. Tan, MD; John S. Thompson, MD; Elizabeth A. Tindall, MD; Robert G. Trapp, MD; Daniel J. Wallace, MD; Walter R. Wallingford, MD; Arthur L. Weaver, MD; Charles W. Weidmann, MD; Gary W. Williams, MD, PhD; Bob I. Wodecki, MD; Suzanne J. Zorn, MD; and Michel Zummer, MD. REFERENCES I. Crofford LJ, Lipsle/PE, Brooks P, et al. Basic biology and clinical application of specific cyclooxygenase-2 inhibitors. Arthritis Rheum. 2000;43:4-13. 2. Dubois RN, Abramson SB, Crofford L, et al. Cyclooxygenase in biology and disease. FASEBJ. 1998;12:10631073. 3. Kivitz A, Eisen G, Zhao WW, et al. Randomized placebocontrolled trial comparing efficacy and safety ofvaldecoxib with naproxen in patients with osteoarthritis.J FaroPract. 2002;51:530-537. 4. Makarowski W, Zhao WW, Bevirt T, Recker DP. Efficacy and safety of the COX-2 specific inhibitor valdecoxib in the management ofosteoarthritis of the hip: A randomized, double-blind, placebo-controlled comparison with naproxen. Osteoarthritis Cartilage. 2002;10:290-296. 5. Sikes DH, Agrawal NM, Zhao WVV,et al. Incidence ofgastroduodenal ulcers associated with valdecoxib compared with that ofibuprofen and diclofenac in patients with osteoarthritis. EurJ GastroenterolHepato[. 2002;14:1101-1111. 6. Goldstein JL, Eisen GM, Agrawal N, et al. Reduced incidence of upper gastrointestinal ulcer complications with the COX-2 selective inhibitor, valdecoxib. Aliment Pharmaco[ Ther. 2004;20:527-538. 7. Eisen GM, Goldstein JL, Hanna DB, Rublee DA. Metaanalysis: Upper gastrointestinal tolerability of valdecoxib, a cyclooxygenase-2-specific inhibitor, compared with nonspecific nonsteroidal anti-inflammatory drugs among patients with osteoarthritis and rheumatoid arthritis. Aliment Pharmaco[ Ther. 2005;21:591-598. 8. Rabeneck L, Goldstein JL, Vu A, et al. Valdecoxib is associated with improved dyspepsia-related health compared with nonspecific NSAIDs in patients with osteoarthritis or rheumatoid arthritis. Amy Gastroentero[. 2005;100:10431050. 220

9. Bensen W, WeaverA, Espinoza L, et al. Efficacy and safety of valdecoxib in treating the signs and symptoms of rheumatoid arthritis: A randomized, controlled comparison with placebo and naproxen. Rheumatology(Oxford). 2002;41 :I 008-I 016. I 0. Pavelka K, Recker DP, Verburg KM. Valdecoxib is as effective as diclofenac in the management of rheumatoid arthritis with a lower incidence ofgastroduodenal ulcers: Results of a 26-week trial. Rheumatology(Oxford). 2003;42: 1207-1215. 1 I. Hippisley-Cox J, Coupland C. Risk of myocardial infarction in patients raking cyclo-oxygenase-2 inhibitors or conventional non-steroidal anti-inflammatory drugs: Population based nested case-control analysis. BMJ.

2005;330:1366. 12. Johnsen SP, Larsson H, Tarone RE, et al. Risk of hospitalization for myocardial infarction among users ofrofecoxib, celecoxib, and other NSAIDs: A population-based case-control study. Arch Intern Med. 2005;165:978-984.

13. Nussmeier NA, Whelton AA, Brown MT, et al. Complications of the COX-2 inhibitors parecoxib and valdecoxib after cardiac surgery. N Engly Med. 2005;352:1081-1091. 14. White WB, Strand V, Roberts R, Whelton A. Effects of the cyclooxygenase-2 specific inhibitor valdecoxib versus nonsteroidal antiinflammatory agents and placebo on cardiovascular thrombotic events in patients with arthritis. Amy Ther. 2004;11:244-250. 15. Ott E, Nussmeier NA, Duke PC, et al, for the Multicenter Study of Perioperative Ischemia (McSPI) Research Group and the Ischemia Research and Education Foundation (IREF) Investigators. Efficacy and safety of the cyclooxygenase 2 inhibitors parecoxib and valdecoxib in patients undergoing coronary artery bypass surgery, y Thorac CardiovascSurg. 2003 ;125:1481-1492. 1 6. Whelton A, Nussmeier NA, Martineau RJ, et al. Safety of parecoxib and valdecoxib in the treatment of postoperative pain following coronary artery bypass graft surgery or major general surgery. Presented at: American College of Cardiology Annual Scientific Session; March 6-9, 2005; Orlando, Fla. 1 7. JenkinsJK, Seligman PJ. Analysis and recommendations for agency action regarding non-steroidal anti-inflammatory drugs and cardiovascular risk. Food and Drug Administration. Available at: http://www.fda.gov/cder/drug/ infopage/COX2/NSAIDdecisionMemo.pdf. Accessed June

20, 2005. 18. Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum. 1988; 31:315-324. 19. Hochberg MC, Chang RW, Dwosh I, et al. The American College of Rheumatology 1991 revised criteria for the classification of global functional status in rheumatoid Volume 28 Number 2

G . W . W i l l i a m s et al.

20.

21.

22.

23.

24.

25.

arthritis. Arthritis Rheum. 1992;35: 498-502. FriesJF. The assessment of disability: From first to future principles. BrJ Rheumatol. 1983;22(SuppI):48-58. Hochberg Y. A sharper Bonferroni procedure for multiple tests of significance. Biometrika. 1988;75:800802. Whelton A, White WB, Kent JD, Verburg KM. Hypertension and edema rates in 6,717 patients for the new COX-2 inhibitor valdecoxib versus the conventional, nonselective NSAIDs and placebo (P-446). Am J Hypertens. 2003;1 6(SuppI):199A. Abstract. Whelton A, Maurath CJ, Verburg KM, Gels GS. Renal safety and tolerability ofcelecoxib, a novel cyclooxygenase-2 inhibitor [published correction appears in AmJ Ther. 2000; 7:341 ]. AmJ Ther. 2000;7:159-175. Whelton A. COX-2-specific inhibitors and the kidney: Effect on hypertension and oedema. J H ypertens Suppl. 2002;20:$31-$35. Williams G, White W, Moore R, et al. Cardiovascular safety of valdecoxib: A meta-analysis of 19 clinical studies in 12,000 patients. Arthritis Rheum.2005;52 (SuppI):S406.

Address correspondence to: Gary W. Williams, MD, PhD, Chairman, Department of Medicine, Scripps Clinic, 10666 North Torrey Pines Road, La Jolla, CA 92037. E-maih [email protected] February 2006

221