3beck.oa

The new england journal of medicine Michael A. Becker, M.D., H. Ralph Schumacher, Jr., M.D., Robert L. Wortmann, M.D., Patricia A. MacDonald, B.S.N., N.P., Denise Eustace, B.A., William A. Palo, M.S., Janet Streit, M.S., and Nancy Joseph-Ridge, M.D.
b a c k g r o u n d
Febuxostat, a novel nonpurine selective inhibitor of xanthine oxidase, is a potential alter- School of Medicine, Chicago (M.A.B.); the native to allopurinol for patients with hyperuricemia and gout.
University of Pennsylvania School of Med-icine, Veterans Affairs Medical Center, Philadelphia (H.R.S.); the University ofOklahoma Department of Medicine, Tulsa We randomly assigned 762 patients with gout and with serum urate concentrations of at least 8.0 mg per deciliter (480 µmol per liter) to receive either febuxostat (80 mg or TAP Pharmaceutical Products, Lake Forest, 120 mg) or allopurinol (300 mg) once daily for 52 weeks; 760 received the study drug.
Ill. (P.A.M., D.E., W.A.P., J.S., N.J.-R.). Ad-dress reprint requests to Dr. Becker at Prophylaxis against gout flares with naproxen or colchicine was provided during weeks 1 through 8. The primary end point was a serum urate concentration of less than 6.0 mg Center, 5841 S. Maryland Ave., Chicago, IL per deciliter (360 µmol per liter) at the last three monthly measurements. The second- 60637, or at mbecker@medicine.bsd.
uchicago.edu.
ary end points included reduction in the incidence of gout flares and in tophus area.
Copyright 2005 Massachusetts Medical Society. The primary end point was reached in 53 percent of patients receiving 80 mg of febux-ostat, 62 percent of those receiving 120 mg of febuxostat, and 21 percent of those re-ceiving allopurinol (P<0.001 for the comparison of each febuxostat group with the allo-purinol group). Although the incidence of gout flares diminished with continuedtreatment, the overall incidence during weeks 9 through 52 was similar in all groups:64 percent of patients receiving 80 mg of febuxostat, 70 percent of those receiving120 mg of febuxostat, and 64 percent of those receiving allopurinol (P=0.99 for 80 mgof febuxostat vs. allopurinol; P=0.23 for 120 mg of febuxostat vs. allopurinol). Themedian reduction in tophus area was 83 percent in patients receiving 80 mg of febux-ostat and 66 percent in those receiving 120 mg of febuxostat, as compared with 50 per-cent in those receiving allopurinol (P=0.08 for 80 mg of febuxostat vs. allopurinol;P=0.16 for 120 mg of febuxostat vs. allopurinol). More patients in the high-dose febux-ostat group than in the allopurinol group (P=0.003) or the low-dose febuxostat groupdiscontinued the study. Four of the 507 patients in the two febuxostat groups (0.8 per-cent) and none of the 253 patients in the allopurinol group died; all deaths were fromcauses that the investigators (while still blinded to treatment) judged to be unrelated tothe study drugs (P=0.31 for the comparison between the combined febuxostat groupsand the allopurinol group).
c o n c l u s i o n s
Febuxostat, at a daily dose of 80 mg or 120 mg, was more effective than allopurinol atthe commonly used fixed daily dose of 300 mg in lowering serum urate. Similar reduc-tions in gout flares and tophus area occurred in all treatment groups.
Downloaded from www.nejm.org on October 31, 2009 . For personal use only. No other uses without permission. Copyright 2005 Massachusetts Medical Society. All rights reserved. f e b u x o s t a t c o m p a r e d w i t h a l l o p u r i n o l i n p a t i e n t s w i t h h y p e r u r i c e m i a a n d g o u t of solubility (about 6.8 mg per deciliter patients
[400 µmol per liter]), is a common biochemical ab- The Febuxostat versus Allopurinol Controlled Trialnormality that reflects supersaturation of the extra- (FACT), a phase 3, randomized, double-blind, 52-cellular fluid with urate and predisposes affected week, multicenter trial, compared the safety andpersons to gout. The clinical manifestations of gout efficacy of febuxostat (taken orally once daily) with(acute gouty arthritis, gouty arthropathy, chronic the safety and efficacy of allopurinol in adult sub-tophaceous gout, uric acid urolithiasis, and gouty jects with gout and with serum urate concentra-nephropathy) result from deposition of monoso- tions of at least 8.0 mg per deciliter (480 µmol perdium urate or uric acid crystals from supersaturated liter). The subjects met the preliminary criteria ofbody fluids.1 The solubility of monosodium urate the American College of Rheumatology for acutein extracellular fluids is influenced by a variety of arthritis of gout.26 The ineligibility criteria includ-factors, including pH, temperature, and sodium ion ed a serum creatinine concentration of more thanand protein concentrations2-9; under certain cir- 1.5 mg per deciliter (133 µmol per liter) or an esti-cumstances, urate solubility may be exceeded at mated creatinine clearance rate of less than 50 ml perconcentrations of 6.0 mg per deciliter (360 µmol minute per 1.73 m2 of body-surface area (becauseper liter) or lower.3 Thus, a major goal in managing allopurinol was included in the study)14,16; preg-gout is long-term reduction of serum urate con- nancy or lactation; use of urate-lowering agents,centrations to clearly subsaturating levels; such re- azathioprine, 6-mercaptopurine, thiazide diuret-duction, if maintained over time, will prevent or ics, or medications containing aspirin (more thanreverse the formation and deposition of urate crys- 325 mg daily) or other salicylates; a body-mass in-tals.10-12 dex (the weight in kilograms divided by the square The most frequently used pharmacologic urate- of the height in meters) of more than 50; a history lowering strategies involve reducing urate produc- of xanthinuria, active liver disease, or hepatic dys-
tion with a xanthine oxidase inhibitor and enhanc- function; use of prednisone at more than 10 mg per
ing urinary excretion of uric acid with a uricosuric day; a change in hormone-replacement therapy or
agent. Urate-lowering agents are limited, however, oral-contraceptive therapy within the previous three
in number, availability, and effectiveness.13 Allo- months; and a history of alcohol abuse or an alco-
purinol, a xanthine oxidase inhibitor, is the most hol intake of more than 14 drinks per week.
commonly prescribed of these agents. The average
dose is 300 mg per day, although dosing recom- study design
mendations range from 100 to 800 mg per day,14-17 We conducted the study at 112 centers in the United
titrated to serum urate15-17 and creatinine clear- States and Canada. Approval was obtained from in-
ance. The side effects of allopurinol, although un- stitutional review boards or independent ethics
common, may be severe or life-threatening and committees. All subjects gave written informed con-
occur more often in patients with renal insuffi- sent and authorization according to the Health In-
ciency.14-17
surance Portability and Accountability Act of 1996.
Febuxostat, a novel, orally administered, non– Subjects already receiving urate-lowering therapy purine analogue inhibitor of xanthine oxidase, is be- underwent a two-week washout period beforeing studied at daily doses of 80 and 120 mg for the undergoing randomization. A computer-generatedmanagement of hyperuricemia in patients with central randomization schedule with a block sizegout. Febuxostat is a potent xanthine oxidase inhib- of three was used to assign each subject to one ofitor, has minimal effects on other enzymes involved three groups: febuxostat (Abbott Laboratories) atin purine and pyrimidine metabolism,18-22 and is 80 mg per day, febuxostat at 120 mg per day, ormetabolized mainly by glucuronide formation and allopurinol (Catalytica Pharmaceuticals) at 300 mgoxidation in the liver.23,24 In a study of subjects per day.
with renal impairment, the serum urate–lowering Initiation of therapy with urate-lowering agents is associated with an increased incidence of acute Downloaded from www.nejm.org on October 31, 2009 . For personal use only. No other uses without permission. Copyright 2005 Massachusetts Medical Society. All rights reserved. The new england journal of medicine gouty attacks10,27,28; accordingly, prophylaxis (250 ing treatment for acute gout flares from weeks 9
mg of naproxen twice daily or 0.6 mg of colchicine through 52.
once daily) was administered to all patients during
the washout period and the first eight weeks of statistical analysis
double-blind treatment. Subsequent flares of gout For the primary efficacy end point, comparisions
were treated at the investigators’ discretion. At two were made sequentially by a two-step closed-testing
weeks and four weeks, and monthly thereafter, each procedure: first, each febuxostat group was com-
patient underwent a physical examination, vital pared with the allopurinol group for noninferiority
signs were recorded, the serum urate concentration by using binomial confidence intervals for the dif-
was measured, renal function was assessed, com- ference between groups; second, each febuxostat
pliance with study drugs was assessed, laboratory group shown to be noninferior to the allopurinol
tests were performed, and concomitant medication group was tested for superiority to the allopurinol
use, gout flares, and adverse events were recorded. group by Fisher’s exact test. Noninferiority to allo-
A treatment-emergent adverse event was defined purinol was declared if the lower bound of the 97.5 as an adverse event occurring during the period be- percent confidence interval was greater than 10 per-tween the first dose and 30 days after the final dose cent. The overall 0.05 alpha level was maintainedof the study drug. A serious adverse event was de- within each step by using binomial 97.5 percentfined as an event that was life-threatening or that confidence intervals for noninferiority tests andresulted in death, hospitalization or prolongation Hochberg’s method for superiority tests.30 Pairwiseof hospitalization, persistent disability or incapac- comparisons with the use of Fisher’s exact test wereity, or a congenital anomaly or birth defect. A treat- also made between the proportions of patients inment-related adverse event was one considered by each treatment group who reached the primary ef-the investigator as possibly, probably, or definitely ficacy end point within each of three groups definedrelated to the study drug. In subjects with tophi, the by baseline urate concentration (less than 9.0 mgarea of one selected tophus was serially measured per deciliter [540 µmol per liter], at least 9.0 butby the following method: two axes through the to- less than 10.0 mg per deciliter [600 µmol per liter],phus at right angles to one another were identified, and 10.0 mg per deciliter or more). Pairwise com-a pen was used to draw marks along the skin on the parisons between groups for the secondary efficacyfirst axis from each side of the tophus until the end points were made with the use of Fisher’s exactnodule obstructed pen movement, the distance be- test for the proportion of subjects with a serumtween the two pen marks over the top of the nodule urate concentration of less than 6.0 mg per deciliterwas measured to the nearest millimeter, and the and the proportion of subjects requiring treatmentprocedure was repeated along the second axis. The for a gout flare from weeks 9 through 52; analysisarea of the tophus was then calculated by multiply- of variance was used to compare the percentage re-ing the two measurements.29 duction from the baseline serum urate concentra-tion; and the Wilcoxon rank-sum test was used to e n d p o i n t s
compare the percentage reduction from baseline to- The primary efficacy end point was a serum urate phus area and number of tophi. All reported P valuesconcentration of less than 6.0 mg per deciliter at are two-sided.
each of the last three monthly measurements. As Post hoc analyses were also performed. Pairwise prespecified, subjects who left the study before comparisons between groups were made with themaking at least three clinic visits were considered use of Fisher’s exact test for the proportions of sub-not to have reached the primary efficacy end point. jects with serum urate concentration of less thanThe secondary efficacy end points included the pro- 5.0 mg per deciliter (300 µmol per liter) and lessportion of subjects with serum urate levels of less than 4.0 mg per deciliter (240 µmol per liter).
than 6.0 mg per deciliter at each visit and the per- Fisher’s exact test and the Wilcoxon rank-sum test,centage reduction from baseline in the serum urate respectively, were used to compare the proportionconcentration at each visit. The clinical end points of subjects requiring treatment for gout flares atwere the percentage reduction from baseline in weeks 49 through 52 and the percentage reductiontophus area, the change in the number of tophi at from baseline tophus area at week 52 between sub-each visit, and the proportion of subjects requir- jects with average post-baseline serum urate con- Downloaded from www.nejm.org on October 31, 2009 . For personal use only. No other uses without permission. Copyright 2005 Massachusetts Medical Society. All rights reserved. f e b u x o s t a t c o m p a r e d w i t h a l l o p u r i n o l i n p a t i e n t s w i t h h y p e r u r i c e m i a a n d g o u t centrations less than 6.0 mg per deciliter and those the subjects had mildly to moderately impaired re-with average concentrations of 6.0 mg or more per nal function (Table 1). Compliance (determined bydeciliter. No adjustments were made to the overall pill count) was similar in all groups (95.0 percent0.05 alpha level for the secondary efficacy end points to 95.5 percent).
or post hoc analyses.
No interim analyses were performed. A sample efficacy
of 750 subjects (250 per group) was targeted to pro- Primary End Point
vide 80 percent power to meet the noninferiority The primary efficacy end point — a serum urate
criteria and 90 percent power to detect a 15 percent concentration of less than 6.0 mg per deciliter at
difference between at least one febuxostat group the last three measurements — was reached by 53
and the allopurinol group for the primary end point, percent of the subjects taking 80 mg of febuxostat,
on the assumption of a true response rate of 60 per- 62 percent of those taking 120 mg of febuxostat, and
cent for allopurinol11,12,31-33 and at least 64 percent 21 percent of those taking allopurinol (P<0.001 for
for febuxostat.
each febuxostat group vs. the allopurinol group) The study was designed by the academic investi- (Table 2). At all ranges of initial urate levels tested, gators and the corporate sponsor (TAP Pharmaceu- the primary end point was reached by higher pro-tical Products). Representatives of TAP collected the portions of febuxostat-treated subjects than allo-data, and statisticians at TAP conducted all statisti- purinol-treated subjects (P<0.001) (Table 2).
cal analyses. All authors had access to the data andvouch for the veracity and completeness of the data Secondary End Pointsand the data analysis. The manuscript was written By week 2 of the study (the first visit after random-in its entirety by the authors.
ization), the proportion of subjects with serumurate concentrations of less than 6.0 mg per decili-ter was significantly higher in the groups receiving febuxostat than in the group receiving allopurinol p a t i e n t c h a r a c t e r i s t i c s
(P<0.001) (Table 2). These differences were sus- Of 1283 subjects screened, 762 were randomly as- tained at all visits through week 52 (P<0.001). Thesigned to treatment (Fig. 1). Of the 762 who under- mean percentage reduction from the baseline se-went randomization, 760 received at least one dose rum urate concentration at the final visit was alsoof the study drug between July 2002 and February greater in both febuxostat groups than in the allo-2004: 256 received 80 mg of febuxostat, 251 received purinol group (Table 2). In addition, post hoc analy-120 mg of febuxostat, and 253 received 300 mg of sis showed that at week 52, the proportions of sub-allopurinol once daily. The mean age, sex ratio, ra- jects with final serum urate concentrations of lesscial distribution, mean baseline serum urate con- than 5.0 or less than 4.0 mg per deciliter were sig-centration, and history or presence of tophi were nificantly greater in both of the febuxostat groupssimilar in the three groups (Table 1). The majority than in the allopurinol group (P<0.001) (Table 2).
of the subjects were white men at least 50 years ofage who reported that they drank alcohol. The sub- Gout Flaresjects had had gout for an average of 12 years, 24 per- During weeks 9 through 52, similar proportions ofcent had tophi or a history of tophi, 16 percent had subjects in each group required treatment for ata history of urolithiasis, and 44 percent had previ- least one gout flare: 64 percent of those receivingously taken allopurinol. Forty-four percent had hy- 80 mg of febuxostat, 70 percent of those receivingpertension, 34 percent had hyperlipidemia, 10 per- 120 mg of febuxostat, and 64 percent of those re-cent had artherosclerotic cardiovascular disease, ceiving allopurinol. During the eight-week prophy-and 62 percent were obese, defined as having a laxis period, a significantly greater proportion ofbody-mass index of 30 or more. The mean baseline subjects receiving 120 mg of febuxostat requiredserum urate concentration ranged from 9.80 to treatment for a gout flare than of those receiving9.90 mg per deciliter (583 to 589 µmol per liter), 80 mg of febuxostat or those receiving allopurinolwith 41 percent of all subjects having a baseline (P<0.001 for both comparisons) (Table 2). With-serum urate concentration of at least 10.0 mg per drawal of prophylaxis was initially accompanied bydeciliter (595 µmol per liter). Thirty-five percent of a markedly increased incidence of gout flares in all Downloaded from www.nejm.org on October 31, 2009 . For personal use only. No other uses without permission. Copyright 2005 Massachusetts Medical Society. All rights reserved. The new england journal of medicine Enrollment
61 Withdrew consent24 Investigator decision59 Other reasons Assignment
Follow-up
Analysis
Figure 1. Flow of Participants through Each Stage of FACT.
UA denotes serum urate concentration.
groups (Fig. 2). The incidence of flare gradually de- By week 52, the median percentage reduction increased thereafter; by weeks 49 through 52, the final tophus area was 83 percent for subjects receivingvisit interval, the incidence was 8 percent among 80 mg of febuxostat, 66 percent for those receivingsubjects receiving 80 mg of febuxostat, 6 percent 120 mg of febuxostat, and 50 percent for those re-among those receiving 120 mg of febuxostat, and ceiving allopurinol. Little change in the number of11 percent among those receiving allopurinol.
tophi over time was noted in any of the treatmentgroups. There were no statistically significant dif- ferences among the groups in the percentage reduc- The percentage reduction in tophus area was as- tion in tophus area or in the reduction in the num-sessed in 156 subjects who had tophi at baseline. ber of tophi (Table 2).
Downloaded from www.nejm.org on October 31, 2009 . For personal use only. No other uses without permission. Copyright 2005 Massachusetts Medical Society. All rights reserved. f e b u x o s t a t c o m p a r e d w i t h a l l o p u r i n o l i n p a t i e n t s w i t h h y p e r u r i c e m i a a n d g o u t Table 1. Baseline Characteristics of the Subjects.*
Febuxostat,
Febuxostat,
Allopurinol,
80 mg/day
120 mg/day
300 mg/day
All Subjects
Variable
P Value†
Baseline serum urate concentration — mg/dl¶ History or presence of tophi — no. of patients (%) Previous urate-lowering therapy — no. of patients (%) Coexisting conditions — no. of patients (%) * Plus–minus values are means ±SD.
† P values were calculated by the chi-square test for categorical variables and by analysis of variance for continuous variables.
‡ Values are based on age at baseline.
§ ¶ To convert values for serum urate to micromoles per liter, multiply by 59.48.
¿ The body-mass index is the weight in kilograms divided by the square of the height in meters.
** The criterion for renal impairment was a calculated creatinine clearance of less than 80 ml per minute per 1.73 m2 of body-surface area, as estimated by the Cockcroft–Gault equation.34 †† Obesity is defined as a body-mass index of 30 or more.
‡‡ A user of low-dose aspirin was defined as a patient who was taking an ongoing total dose of 325 mg per day or less at completion of the study.
§§ The criteria for the metabolic syndrome were a serum triglyceride level of at least 150 mg per deciliter (1.7 mmol per liter), a blood pressure of at least 130/85 mm Hg, and a fasting serum glucose level of at least 110 mg per deciliter (6.1 mmol per liter).
jects requiring treatment for a gout flare was lower A post hoc analysis of the results of the trial was among subjects who reached a mean post-baselineperformed to test for differences in the reduction serum urate concentration of less than 6.0 mg perof gout flares and tophus area between subjects deciliter than among those who did not (6 percentwith a mean post-baseline serum urate concentra- vs. 14 percent, P=0.005). The median reductiontion of less than 6.0 mg per deciliter and those with from baseline in tophus area at week 52 was 75 per-a concentration of 6.0 mg or more per deciliter. Dur- cent among subjects who reached an average post-ing weeks 49 through 52, the proportion of sub- baseline serum urate concentration of less than Downloaded from www.nejm.org on October 31, 2009 . For personal use only. No other uses without permission. Copyright 2005 Massachusetts Medical Society. All rights reserved. The new england journal of medicine Table 2. Primary and Secondary End Points.*
Febuxostat,
Febuxostat,
Allopurinol,
End Point
80 mg/day
120 mg/day
300 mg/day
Primary end point
Serum urate <6.0 mg/dl at last 3 monthly visits† Difference in proportions, 80-mg febuxostat vs. allopurinol — % Difference in proportions, 120-mg febuxostat vs. allopurinol — % Secondary end points
Percent change in serum urate concentration from baseline at final visit Serum urate <6.0 mg/dl at last 3 visits, according to baseline concen- 6.0 mg per deciliter, as compared with 50 percent were mild to moderate in severity. The incidence ofamong those who did not (P=0.06).
serious adverse events was similar in all groups;serious adverse events occurred in 51 subjects, 34 a d v e r s e e v e n t s
of whom continued in the study while the event re- The incidence of adverse events was similar in the solved without recurrence. Four of the 507 patientsthree treatment groups (Table 3). Treatment-related in the two groups receiving febuxostat (0.8 percent)adverse events included abnormal liver-function test and none of the 253 in the allopurinol group died;results, diarrhea, headaches, joint-related signs and all deaths were judged by the investigators to be un-symptoms, and musculoskeletal and connective- related to the study drugs. The difference betweentissue signs and symptoms. Most adverse events the numbers of deaths in the febuxostat groups Downloaded from www.nejm.org on October 31, 2009 . For personal use only. No other uses without permission. Copyright 2005 Massachusetts Medical Society. All rights reserved. f e b u x o s t a t c o m p a r e d w i t h a l l o p u r i n o l i n p a t i e n t s w i t h h y p e r u r i c e m i a a n d g o u t Table 2. (Continued.)
Febuxostat,
Febuxostat,
Allopurinol,
End Point
80 mg/day
120 mg/day
300 mg/day
No. of patients (median % change in area) No. of patients (median change in no. of tophi/patient) Post hoc analysis of serum urate at final visit * Four subjects (one receiving 80 mg of febuxostat, one receiving 120 mg of febuxostat, and two receiving allopurinol) were excluded, as prespecified, from the efficacy analysis because their baseline serum urate concentration on day 2 was less than 8.0 mg per deciliter.
† To convert values for serum urate to micromoles per liter, multiply by 59.48.
‡ The 97.5 percent confidence interval (CI) based on the normal approximation for the binomial distribution is given.
§ P values were calculated by Fisher's exact test. An overall alpha level of 0.05 was maintained within each step by using Hochberg's method for superiority30 and a binomial 97.5 percent confidence interval for noninferiority.
¶ The difference was statistically significant for the comparison with allopurinol.
¿ The difference was statistically significant for the comparison with 120 mg per day of febuxostat.
and the allopurinol group was not statistically sig- which accounted for the withdrawal of five patientsnificant (P=0.31). There were two deaths in the receiving 80 mg of febuxostat, seven receivinggroup receiving 80 mg of febuxostat: one from con- 120 mg of febuxostat, and one receiving allopuri-gestive heart failure and respiratory failure in a nol (P=0.04 for the comparison between the 120-65-year-old man, and one from retroperitoneal mg febuxostat and the allopurinol groups). Fourbleeding ascribed to anticoagulation therapy in a subjects receiving 80 mg of febuxostat, four receiv-77-year-old man. Two deaths occurred in the group ing 120 mg of febuxostat, and one receiving allo-receiving 120 mg of febuxostat: one from metastat- purinol discontinued the study because of rashes.
ic colon cancer in a 74-year-old man, and one from Most of these were localized and transient maculo-cardiac arrest in a 68-year-old man.
papular rashes that occurred during prophylactic Eighty-eight subjects in the 80-mg febuxostat treatment with either colchicine or naproxen and group, 98 in the 120-mg febuxostat group, and resolved after topical treatment.
66 in the allopurinol group discontinued the study(P=0.003 for the comparison between the 120-mg febuxostat and the allopurinol groups) (Fig. 1).
The most frequent reasons for discontinuation were This large, randomized, controlled clinical trial,lost to follow-up, adverse events, and gout flares. conducted in subjects with hyperuricemia and gout,The most common adverse event leading to with- compared treatment with febuxostat and allopuri-drawal was abnormal liver-function test results, nol with regard to safety, urate-lowering efficacy, Downloaded from www.nejm.org on October 31, 2009 . For personal use only. No other uses without permission. Copyright 2005 Massachusetts Medical Society. All rights reserved. The new england journal of medicine Flare in a 4-Wk Period
Percentage of Subjects with at Least One
Figure 2. Subjects Requiring Treatment for Gout Flares.
The percentage of subjects in each interval is calculated by dividing the number of subjects with at least one gout flare in that interval by the number of subjects exposed to at least one dose of drug in that interval. Subjects may be counted in more than one interval. The subjects received prophylaxis during the period from day 1 to week 8. The results for the 80-mg febuxostat group are shown in blue, those for the 120-mg febuxostat group in pink, and those for the allopurinol group in yellow.
incidence of gout flares, and changes in tophus area. ference between the febuxostat and the allopurinolAdministration of febuxostat or allopurinol result- groups was not statistically significant (P=0.31).
ed in prompt (within two weeks) and persistent re- Long-term studies are ongoing to provide furtherduction in serum urate concentration; however, all evaluation of the safety profile of febuxostat.
urate-lowering end points requiring serum urate The high rate of gout flares in all groups during concentrations of less than 6.0 mg per deciliter were prophylaxis, and especially after withdrawal of pro-reached by significantly greater proportions of sub- phylaxis, calls attention to a well-described24,28,33jects receiving daily febuxostat (80 or 120 mg) than paradox with important implications for successfulsubjects receiving allopurinol (300 mg). The clini- management of gout: the risk of acute gout flares iscal outcomes (reduction in gout flares and in tophus increased early in the course of urate-lowering treat-area) were not different in the febuxostat and allo- ment. This study clearly documents a role for morepurinol groups.
sustained prophylaxis during the initiation of urate- In this trial, the overall incidences of treatment- lowering therapy than was provided here.
related adverse events were similar for all treatment Our study was designed to test the hypothesis groups, and most were mild to moderate in severity. that febuxostat is not inferior to allopurinol withThe rates of discontinuation were similar in the respect to urate-lowering efficacy. On the basis of80-mg febuxostat and the allopurinol groups but published studies,11,12,31-33 we predicted that thewere significantly higher in the 120-mg febuxostat primary end point (a serum urate concentration ofgroup than in the other two groups (P=0.003). The less than 6.0 mg per deciliter) would be reached byhigher rate of discontinuation in the 120-mg febux- 50 percent to 60 percent of the subjects receivingostat group was due to the higher incidence of gout allopurinol at a dose of 300 mg per day. In fact, onlyflares and adverse events in this group. No serious 21 percent reached this end point. Two factorsrashes or hypersensitivity reactions occurred in this might contribute to the lower-than-expected urate-study. There were four deaths in the febuxostat lowering efficacy of allopurinol. First, study entrygroups and none in the allopurinol group; the dif- required a baseline serum urate concentration of at Downloaded from www.nejm.org on October 31, 2009 . For personal use only. No other uses without permission. Copyright 2005 Massachusetts Medical Society. All rights reserved. f e b u x o s t a t c o m p a r e d w i t h a l l o p u r i n o l i n p a t i e n t s w i t h h y p e r u r i c e m i a a n d g o u t Table 3. Summary of Adverse Events.
Febuxostat,
Febuxostat,
Allopurinol,
80 mg/day
120 mg/day
300 mg/day
Adverse Event
Most frequent treatment-related adverse events¿ Joint-related signs and symptoms (arthralgia, joint stiffness or swelling) Musculoskeletal and connective-tissue signs and symptoms (back, chest- wall, flank, or extremity pain and musculoskeletal stiffness) Gastrointestinal atonic and hypomotility disorders (constipation, gastro- Neurologic signs and symptoms (dizziness, dysgeusia) Gastrointestinal signs and symptoms (epigastric and stomach discomfort) * A treatment-emergent event was an adverse event that occurred during the period from the first dose to 30 days after the † The difference from the allopurinol group was significant (P=0.01) by Fisher’s exact test.
‡ A serious adverse event was life-threatening or resulted in death, hospitalization or prolongation of hospitalization, per- sistent or significant disability or incapacity, or congenital anomaly or birth defect. All of these events were considered by the investigator to be not related or unlikely to be related to the study drug.
§ There were no statistically significant differences among the groups.
¶ A treatment-related adverse event was considered by the investigator to be possibly, probably, or definitely related to the ¿ Adverse events were classified according to the definitions in the Medical Dictionary for Regulatory Activities (MedDRA)35 on the basis of the signs and symptoms reported by the investigators. The most frequent treatment-related adverse events were defined as those reported for at least 2 percent of the subjects in at least one of the treatment groups.
least 8.0 mg per deciliter, and the mean baseline more, no clinical trials have been conducted to as-serum urate concentration was nearly 10.0 mg per sess the safety and efficacy of titration of the dosedeciliter, a level exceeded by 41 percent of the sub- of allopurinol according to serum urate levels.
jects. These baseline levels may not be uncommon In retrospective, nonrandomized studies and in in the current population of patients with gout,12 small, prospective studies, attainment and mainte-but they exceed those reported several decades ago, nance of serum urate concentrations of less thanwhen allopurinol was introduced.36,37 Second, in 6.0 mg per deciliter have been associated with long-order to confirm the persistence of the urate-low- term benefits in patients with gout, including re-ering effect, the primary end point was defined as duction in the frequency of gout flares and decreasethree successive measurements of serum urate of in the size or number of tophi.10-12 In this study,less than 6.0 mg per deciliter. It is likely that allopu- reductions in the incidence of gout flares and inrinol would have been more effective at lowering tophus area (the clinical end points) were also ob-urate levels if the dose had been titrated as recom- served over time and were similar in all treatmentmended in the allopurinol package insert. In this groups. However, the current study was only 52trial, however, titration of allopurinol would have weeks in duration, and post hoc analysis of thecompromised the blinding of the study. Further- relation between the incidence of gout flares and Downloaded from www.nejm.org on October 31, 2009 . For personal use only. No other uses without permission. Copyright 2005 Massachusetts Medical Society. All rights reserved. The new england journal of medicine an average post-baseline serum urate concentra- acute and the chronic manifestations of gout. Sec-tion of less than 6.0 mg per deciliter or 6.0 mg or ond, the greater reduction in gout flares and to-more per deciliter found a significant difference phus area over time when the serum urate con-only in the last 4 weeks. This suggests that a longer centration is maintained at less than 6.0 mg pertrial would be necessary to distinguish between deciliter supports the use of the subsaturatingurate-lowering agents with regard to superiority in range of less than 6.0 mg per deciliter as an appro-clinical outcome.
priate target for the management of symptomatic The results of this study provide information hyperuricemia.
generally applicable to the management of hyper- Dr. Becker, Dr. Schumacher, and Dr. Wortmann report serving as consultants for TAP Pharmaceutical Products. Dr. Joseph-Ridge, uricemia in patients with gout. First, sustained Ms. MacDonald, Ms. Eustace, Ms. Streit, and Mr. Palo are employ- lowering of serum urate was accompanied over ees of TAP Pharmaceutical Products.
months by a reduction in the incidence of gout We are indebted to Kazutaka Shiobara, Satoru Hoshide, Yasuhiro Takahashi, and Barbara Hunt for their review of the manuscript and flares and in tophus area, confirming the beneficial to Susan Cazzetta for her assistance in the preparation of the manu- effects of sustained urate reduction on both the script. a p p e n d i x
The principal investigators in the Febuxostat versus Allopurinol Controlled Trial (FACT) are as follows: D.M. Aboud (El Paso, Tex.), T.C.
Adamson III (San Diego, Calif.), C.G. Andersen (Salt Lake City), J.D. Angeloni (Bala Cynwyd, Pa.), A.B. Aven (Arlington Heights, Ill.), R.F.
Bader (Santa Ana, Calif.), A.R. Baldassare (St. Louis), H.S.B. Baraf (Wheaton, Md.), S.A. Bart, Sr. (Gainesville, Fla.), M.A. Becker (Chicago),C. Birbara (Worcester, Mass.), B.I. Blatt (Havertown, Pa.), S. Bookbinder (Ocala, Fla.), D.T. Borkert (Lakewood, Colo.), D.W. Bouda (Omaha,Nebr.), B.T. Bowling (Endwell, N.Y.), S.S. Brady (Naples, Fla.), M.L. Brandon (San Diego, Calif.), F.X. Burch (San Antonio, Tex.), R. Cattan(Miami), C.M. Chappel (Kissimmee, Fla.), W.F. Chase (Austin, Tex.), P. Chatpar (Plainview, N.Y.), A. Cividino (Hamilton, Ont., Canada),D.H. Cohen (Hewlett, N.Y.), J.J. Cohen (Davie, Fla.), G.V. Collins (Charlotte, N.C.), R.L. Collins (Columbia, S.C.), G.A. Colner (Oceanside,Calif.), A. Dahdul (Springfield, Mass.), G. DiVittorio (Mobile, Ala.), D. Doolin (Edgewater, Fla.), R.K. Dore (Anaheim, Calif.), W. Drum-mond (Dallas), B. Feingold (Manhasset, N.Y.), J.J. Fiechtner (Lansing, Mich.), C.L. Fisher, Jr. (Newport News, Va.), D. Fitz-Patrick (Hono-lulu), F.D. Fraser (Stoney Creek, Ont., Canada), D.L. Fried (Warwick, R.I.), F. Galef (Vista, Calif.), R.E. Gaona (San Antonio, Tex.), N.B. Gay-lis (Aventura, Fla.), S.L. Glickstein (Minneapolis), J.E. Greenwald (St. Louis), J.S. Grober (Evanston, Ill.), C.S. Guy (Creve Coeur, Mo.),J. Habros (Scottsdale, Ariz.), D. Haselwood (Fair Oaks, Calif.), J.R. Hill (Broomfield, Colo.), S. Hole (Edgewater, Fla.), P.A. Holt (Balti-more), J.P. Huff (San Antonio, Tex.), R.T. Huling (Olive Branch, Miss.), T. Isakov (Lyndhurst, Ohio), A.M. Jackson (Bartlett, Tenn.), A.T.
Kaell (Port Jefferson Station, N.Y.), S.P. Kafka (Duncansville, Pa.), L.G. Karlock (Austintown, Ohio), R.M. Karr (Everett, Wash.), R.S. Kauf-mann (Austell, Ga.), A. Kelly (Edmonton, Alta., Canada), J.D. King (Selmer, Tenn.), L.C. Kirby II (Mesa, Ariz.), M. Kohen (Port Orange,Fla.), A.R. Kuhn (St. Petersburg, Fla.), B. Lasko (Toronto), D. Lewis (Little Rock, Ark.), T.W. Littlejohn III (Winston-Salem, N.C.), B. Long(Cleveland), H. Luque (Los Angeles), R.D. Madder (Beaver, Pa.), H.W. Marker (Memphis, Tenn.), P.D. Matz (Medford, Oreg.), H.H. McIl-wain (Tampa, Fla.), B.K. McLean (Birmingham, Ala.), C. Mendoza (Toms River, N.J.), C. Multz (San Jose, Calif.), C.D. Okonski (St. Joseph,Mich.), W.R. Palmer (Omaha, Nebr.), J.E. Pappas (Lexington, Ky.), A.J. Pareigis (Moline, Ill.), R.Z. Paster (Oregon, Wis.), N.R. Patel (Ketter-ing, Ohio), R.A. Petrus II (Tampa, Fla.), B.C. Pogue (Boise, Idaho), A. Porges (Hewlett, N.Y.), R.W. Powell (Newark, Del.), H.M. Prupas(Reno, Nev.), K. Raben (South Miami, Fla.), B.G. Rankin (DeLand, Fla.), L.R. Rocamora (Winston-Salem, N.C.), M.A. Rosemore (Huey-town, Ala.), S. Rosenblatt (Irvine, Calif.) J. Rubino (Raleigh, N.C.), G.E. Ruoff (Kalamazoo, Mich.), B.S. Samuels (Dover, N.H.), J. Schecht-man (Glendale, Ariz.), H.R. Schumacher (Philadelphia), E.A. Sheldon (Miami), W.J. Shergy (Huntsville, Ala.), D. Shu (Coquitlam, B.C.,Canada), I.J. Siegel (Markham, Ont., Canada), E.J. Spiotta, Jr. (Memphis, Tenn.), J. Tesser (Phoenix, Ariz.), A. Torres (St. Petersburg, Fla.),S. Touger (Birmingham, Ala.), R.G. Trapp (Springfield, Ill.), Q.H. Usmani (Toms River, N.J.), M.A. Vacker (Davie, Fla.), R.M. Valente (Lin-coln, Nebr.), N. Wei (Frederick, Md.), C.W. Wiesenhutter (Coeur d’Alene, Idaho), H.T. Williams (Birmingham, Ala.), S.M. Wolfe (Dayton,Ohio), L.K. Wright (Birmingham, Ala.), and H. Zaharowitz (St. Petersburg, Fla.).
r e f e r e n c e s
10. Shoji A, Yamanaka H, Kamatani N.
mia. Curr Opin Rheumatol 2002;14:281-6.
Klinenberg J, Whitehouse M. Urate binding: A retrospective study of the relationship a clue to the pathogenesis of gout. J Rheu- between serum urate level and recurrent at- tacks of gouty arthritis: evidence for reduc- Sledge CB, eds. Kelley’s textbook of rheu- tion of recurrent gouty arthritis with anti- matology. 6th ed. Philadelphia: W.B. Saun- ies on the nucleation of monosodium urate at 37 degrees C. Arthritis Rheum 1980;23: Allen DJ, Milosovich G, Mattocks AM.
11. Perez-Ruiz F, Calabozo M, Pijoan J, Her-
Inhibition of monosodium urate needle crys- Perl-Treves D, Addadi L. A structural ap- rero-Beites AM, Ruibal A. Effect of urate- tal growth. Arthritis Rheum 1965;8:1123-33.
proach to pathological crystallizations — lowering therapy on the velocity of size re- gout: the possible role of albumin in sodium duction of tophi in chronic gout. Arthritis on the solubility of monosodium urate. Ar- urate crystallization. Proc R Soc Lond B Biol 12. Perez-Ruiz F, Alonso-Ruiz A, Calabozo
Kippen I, Klinenberg JR. Solubility of uric ies of urate crystallisation in relation to Ruiz-Lucea E. Efficacy of allopurinol and benzbromarone for the control of hyperuri- caemia: a pathogenic approach to the treat- Downloaded from www.nejm.org on October 31, 2009 . For personal use only. No other uses without permission. Copyright 2005 Massachusetts Medical Society. All rights reserved. f e b u x o s t a t c o m p a r e d w i t h a l l o p u r i n o l i n p a t i e n t s w i t h h y p e r u r i c e m i a a n d g o u t 22. Becker MA, Kisicki J, Khosraven R, et al.
five years in 208 gouty subjects. Ann Intern Febuxostat (TMX-67), a novel, non-purine, 13. Schlesinger N, Schumacher HR Jr. Gout:
selective inhibitor of xanthine oxidase, is 29. Schumacher HR, Becker MA, Palo W,
safe and decreases serum urate in healthy Hunt B, MacDonald PA, Joseph-Ridge N.
volunteers. Nucleosides Nucleotides Nucle- Tophaceous gout — quantitative evaluation 14. Schlesinger N. Management of acute
by direct physical measurement. J Rheuma- and chronic gouty arthritis: present state-of- 23. Hoshide S, Nishimura S, Ishii S, Mat-
suzawa K, Saito N, Tanaka T. Metabolites of 30. Hochberg Y. A sharper Bonferroni pro-
15. Terkeltaub RA. Gout. N Engl J Med
TMX-67, a new pharmaceutical entity for the cedure for multiple tests of significance.
16. Fam AG. Gout in the elderly: clinical
their pharmacokinetic profiles in humans.
31. Brewis I, Ellis RM, Scott JT. Single daily
24. Khosravan R, Mayer M, Grabowski B,
17. Harris MD, Siegel LB, Alloway JA. Gout
Vernillet L, Wu J-T, Joseph-Ridge N. Febux- 32. Rodnan GP, Robin JA, Tolchin SF, Elion
and hyperuricemia. Am Fam Physician 1999; ostat, a novel non-purine selective inhibitor of xanthine oxidase — effect of mild and efficacy of a single daily dose. JAMA 1975; 18. Okamoto K, Eger BT, Nishino T, Kondo
S, Pai EF, Nishino T. An extremely potent in- cokinetics, pharmacodynamics, and safety.
33. Yamanaka H, Togashi R, Hakoda M, et
hibitor of xanthine oxidoreductase: crystal Arthritis Rheum 2004;50:S337. abstract.
al. Optimal range of serum urate concentra- structure of the enzyme-inhibitor complex 25. Mayer MD, Khosravan R, Vernillet L, Wu
tions to minimize risk of gouty attacks dur- JT, Joseph-Ridge N, Mulford DJ. Pharmaco- ing anti-hyperuricemic treatment. Adv Exp 19. Takano Y, Hase-Aoki K, Horiuchi H, et al.
ostat, a new non-purine selective inhibitor 34. Cockcroft DW, Gault MH. Prediction of
Selectivity of febuxostat, a novel non-purine of xanthine oxidase, in subjects with renal creatinine clearance from serum creatinine.
inhibitor of xanthine oxidase/xanthine de- impairment. Am J Ther 2005;12:22-34.
hydrogenase. Life Sci 2005;76:1835-47.
26. Wallace SL, Robinson H, Masi AT,
35. Brown EG, Wood L, Wood S. The
20. Horiuchi H, Ota M, Kobayashi M, et al.
Decker JL, McCarty DJ, Yu TF. Preliminary medical dictionary for regulatory activities criteria for the classification of the acute activity and potency in renal xanthine calcu- arthritis of primary gout. Arthritis Rheum 36. Campion EW, Glynn RJ, DeLabry LO.
lus formation of two xanthine oxidase/xan- Asymptomatic hyperuricemia: risks and con- 27. Kot TV, Day RO, Brooks PM. Preventing
sequences in the Normative Aging Study.
acute gout when starting allopurinol ther- apy: colchicine or NSAIDs? Med J Aust 1993; 37. Hall AP, Barry PE, Dawber TR, Mc-
21. Osada Y, Tsuchimoto M, Fukushima H,
et al. Hypouricemic effect of the novel xan- 28. Yu TF, Gutman AB. Efficacy of colchi-
peruricemia: a long-term population study.
thine oxidase inhibitor, TEI-6720, in rodents.
cine prophylaxis in gout: prevention of re- current gouty arthritis over a mean period of Copyright 2005 Massachusetts Medical Society. physician-journalist
The Journal is seeking a physician with substantial reporting experience to write articles on timely topics in medicine and society for the Perspective section. Send curriculum vitae and writing samples to Perspective Editor, New England Journal of Medicine, 10 Shattuck St., Boston, MA 02115, or at writer@nejm.org.
Downloaded from www.nejm.org on October 31, 2009 . For personal use only. No other uses without permission. Copyright 2005 Massachusetts Medical Society. All rights reserved. CORRECTION
watchful as new agents are introduced into practice.4 Vigilance and post-marketing pharmacoepidemiology can be particularly enlighten- Febuxostat versus Allopurinol for Gout
To the Editor: The article by Becker et al. (Dec. 8 issue)1 includes an Johns Hopkins University School of Medicine inaccurate and misleading statement regarding the comparison of 80 mg per day of febuxostat with allopurinol for gout. The authors state that ``the rates of discontinuation were similar in the 80-mg febuxostat and the allopurinol groups but were significantly higher in the 120-mg References
febuxostat group than in the other two groups (P=0.003).´´ The authors do not present a statistical analysis comparing the rates 1. Lawrence RC, Helmick CG, Arnett FC, et al. Estimates of the of discontinuation in the 80-mg febuxostat group with those in the al- prevalence of arthritis and selected musculoskeletal disorders in lopurinol group. On the basis of the data they present, there was a the United States. Arthritis Rheum 1998;41:778-799.
significantly higher rate of discontinuation in the group receiving 80 mg per day of febuxostat (P=0.04 by Fisher’s exact test).
2. Kuncl RW, Duncan G, Watson D, Alderson K, Rogawski MA, Peper M. Colchicine myopathy and neuropathy. N Engl J Med This result affects the conclusions of the authors. A higher discontin- uation rate in the group receiving 80 mg per day of febuxostat implies that febuxostat was not as well tolerated as allopurinol. Febuxostat 3. Singer JZ, Wallace SL. The allopurinol hypersensitivity syndrome: may be an advance in the treatment of gout, but we need to be clear unnecessary morbidity and mortality. Arthritis Rheum 1986;29:82- and precise in interpreting the trial data regarding its use.
4. Drazen JM. COX-2 inhibitors – a lesson in unexpected problems.
The authors reply: Our article contains an inaccuracy affecting in- terpretation of the study data. As identified by Dr. Lustberg, when References
Fisher’s exact test is used to compare rates of discontinuation in the 80-mg febuxostat and allopurinol groups, a statistically significant dif- 1. Becker MA, Schumacher HR Jr, Wortmann RL, et al. Febuxostat ference (P=0.04) between the two groups is found. Therefore, the compared with allopurinol in patients with hyperuricemia and gout.
statement regarding premature discontinuation should read, ``The rates of discontinuation were significantly higher in both the 80-mg febuxostat group and the 120-mg febuxostat group than in the allop- urinol group (P=0.04 and P=0.002, respectively).´´ Corrections should To the Editor: Given the incapacitating nature of acute attacks of gout be noted for similar text that appears in the Abstract and in the dis- and their substantial prevalence in the United States,1 the need to cussion of adverse events in the Results section of our article.
bring new, safer, and more effective agents to market is a priority.
Moreover, clinicians regularly face the quandary that existing forms We reviewed the basis of these differences. Results reported in the of therapy (including nonsteroidal antiinflammatory drugs, colchicine, article for comparisons of groups that were relevant to premature and allopurinol) pose a risk of meaningful toxic effects,2,3 especially discontinuation were those determined with the use of a continuity- among persons with the greatest need for treatment, such as the el- adjusted chi-square test (P=0.053 for comparison of the allopurinol derly and those with chronic renal insufficiency.
group with the 80-mg febuxostat group, and P=0.003 for comparison of the allopurinol group with the 120-mg febuxostat group), rather than As such, the arrival of febuxostat is greatly anticipated. Its superior those determined with a Fisher’s exact test, as intended.
efficacy as compared with allopurinol in the reduction of serum urate concentrations, even to optimally low levels, is heralded in the Journal.
All other analyses in the article have been rechecked, and an addi- Caution, however, needs to be exercised inasmuch as the reported tional point for correction has been identified. In Table 1 of the article, frequency of adverse events leading to discontinuation of the drug oc- data about renal impairment are based on calculated creatinine clear- curred two and three times as often in the low-dose and high-dose ance, and the P value should be 0.26, not 0.90. The P value of 0.90 febuxostat groups, respectively, as in the allopurinol group. More- was based on renal impairment as defined in the ineligibility crite- over, the occurrence of four deaths in the febuxostat groups, as com- ria that were outlined in the Methods section. We believe that these pared with none in the allopurinol group, is further reason for pause.
changes do not affect the overall conclusion of the article, which is A compelling recent lesson regarding new arthritis medication is to be that febuxostat at a dose of 80 mg or 120 mg daily is more effective Downloaded from www.nejm.org on October 31, 2009 . For personal use only. No other uses without permission. Copyright 2005 Massachusetts Medical Society. All rights reserved. than allopurinol at a dose of 300 mg daily in lowering serum urate in We thank Dr. Gelber for calling attention to the data in Figure 1 of the original article showing increased rates of premature discontinu- ation among patients treated with febuxostat. Although rashes and abnormal results of liver-function tests — the major adverse reactions leading to withdrawal — were mild to moderate in severity and re- versible after discontinuation of febuxostat, we agree that roles for vigilance and post-marketing pharmacoepidemiology are essential in establishing the ultimate safety profile for febuxostat and, indeed, any University of Chicago Pritzker School of Medicine Downloaded from www.nejm.org on October 31, 2009 . For personal use only. No other uses without permission. Copyright 2005 Massachusetts Medical Society. All rights reserved.

Source: http://www.podiatry.com/images/eZines/PracticePerfect/167/Uloric%20study.pdf