Clinical Therapeutics/Volume 32, Number 14, 2010

Effect of Prophylaxis on Gout Flares After the Initiation of

Urate-Lowering Therapy: Analysis of Data From Three Phase
III Trials
Robert L. Wortmann, MD1; Patricia A. MacDonald, RN, NP2; Barbara Hunt, MS2; and
Robert L. Jackson, MD, MBA2
1

Section of Rheumatology, Dartmouth Medical School, Lebanon, New Hampshire; and 2Takeda Global
Research and Development Center, Inc, Deerfield, Illinois

ABSTRACT
Background: Use of urate-lowering therapy (ULT),
such as febuxostat or allopurinol, is recommended
for the long-term management of hyperuricemia in
patients with gout to reduce the incidence of acute
flares. Because of the paradoxical relationship between early use of ULT and the increased incidence
of gout flares, prophylaxis with either low-dose colchicine or NSAIDs has been recommended, although
there have been concerns about the long-term prophylactic use of these agents.
Objectives: The present analysis examined flare
rates during the 3 Phase III trials of febuxostat based
on mean postbaseline serum urate (sUA) concentrations and duration of prophylaxis. Adverse events
(AEs) were assessed by prophylaxis with colchicine or
naproxen.
Methods: This investigator-initiated, post hoc reanalysis of data on gout flares from the 3 randomized, placebo-controlled, Phase III trials evaluated the proportion of
patients requiring treatment for gout flares at 4-week intervals based on mean postbaseline sUA concentrations
6.0 and 6.0 mg/dL. The 3 trials enrolled males or
females aged 18 85 years who had a diagnosis of gout
and a baseline sUA concentration 8.0 mg/dL. Patients
received ULT (febuxostat or allopurinol) or placebo for 6
months or 1 year and flare prophylaxis with colchicine
0.6 mg/d or naproxen 250 mg BID for 8 weeks or 6
months. The prophylactic regimen was chosen at the discretion of the investigator, based on renal function and
known intolerance to either drug. Patients with an estimated creatinine clearance 50 mL/min were not to receive naproxen. AEs were summarized based on prophylaxis with colchicine or naproxen.
Results: The 3 trials enrolled a total of 4101 patients
with gout. The majority were white (80.1%), male
(94.5%), and obese (body mass index 30 kg/m2)

2386

(62.8%). The mean duration of gout ranged from

10.9 11.9 years, and the mean baseline sUA concentration ranged from 9.6 9.9 mg/dL. Flare rates increased sharply (up to 40%) at the end of 8 weeks of
prophylaxis and then declined gradually, whereas flare
rates were consistently low (range, 3%5%) at the end
of 6 months of prophylaxis. Mean postbaseline sUA
concentrations were correlated with flare rates; by the
end of each study, patients with a mean postbaseline
sUA concentration 6.0 mg/dL had fewer flares than
did those with a mean postbaseline sUA concentration
6.0 mg/dL. There were differences in rates of AEs
between prophylaxis groups, but the rates did not increase with increased duration of prophylaxis.
Conclusion: This analysis of gout flare data from the
3 Phase III trials of febuxostat found that flare prophylaxis for up to 6 months during the initiation of ULT
appeared to provide greater benefit than flare prophylaxis for 8 weeks, with no increase in AEs. (Clin Ther.
2010;32:2386 2397) 2010 Elsevier HS Journals,
Inc.
Key words: allopurinol, colchicine, febuxostat, flares,
gout, naproxen.

INTRODUCTION
Urate-lowering therapy (ULT) is indicated for individuals with hyperuricemia (serum urate [sUA] concentration 6.8 mg/dL) and gout who have had multiple
acute flares and/or developed such complications as
tophaceous deposits.1 Acute gout flares are known to
occur as a consequence of the initiation of ULT and
Accepted for publication November 8, 2010.
doi:10.1016/j.clinthera.2011.01.008
0149-2918/$ - see front matter
2010 Elsevier HS Journals, Inc. All rights reserved.

Volume 32 Number 14

R.L. Wortmann et al.

effective reduction of sUA concentrations.2 In fact, the
greater the reduction in sUA, the more likely a flare.3
This paradox of gout treatment was noted during the
clinical development of uricosuric agents and allopurinol, generally in uncontrolled trials involving small
numbers of patients,4 8 and was confirmed in the febuxostat clinical trial program.9 11
To help prevent flares, it is recommended that initiation of ULT be accompanied by low-dose colchicine
or NSAIDs.1,12 Before the febuxostat clinical trial program, the limited amount of data on flare prophylaxis
from clinical trials left concerns about chronic use of
these agents. Only 2 randomized controlled trials that
examined the benefits and tolerability of colchicine
prophylaxis during 6 months of ULT with allopurinol13 and probenecid14 reported significant reductions
in the frequency and severity of acute attacks in patients receiving colchicine compared with those receiving ULT only. In the allopurinol/colchicine study, in
which colchicine 0.6 mg was administered BID, overall
rates of acute gout flares were 33% in the allopurinol/
colchicine group and 77% in the allopurinol/placebo
group (P 0.008).13 The only significant difference in
adverse events (AEs) was an increase in diarrhea
among patients receiving colchicine compared with
those receiving allopurinol alone. In the other study, in
which probenecid was administered with colchicine
0.5 mg TID, the annual rate of acute gout flares was 2.3
in the probenecid/colchicine group and 6.0 in the probenecid/placebo group (P 0.05).14 Patients receiving
colchicine had higher rates of gastrointestinal AEs than
did those who received probenecid alone.14
Febuxostat is a selective xanthine oxidase inhibitor15
that is approved for the treatment of hyperuricemia in
patients with gout in the United States, Canada, South
Korea, and Europe.16 Three double-blind clinical trials
compared the use of febuxostat and allopurinol for up to
1 year in a total of 4000 patients and provided evidence
regarding the relationship between gout flares, use of
ULT, and use of flare prophylaxis.9 11 Although these
trials were not designed to assess different prophylactic
regimens in a blinded, controlled manner, they were the
largest trials of ULT to examine the impact of both colchicine and the NSAID naproxen on flare rates over different durations of prophylaxis. In addition, 2 open-label
extension trials examined the benefits of long-term ULT
(up to 5 years) on flare rates and found that use of ULT
for 1 year was associated with near-elimination of gout
flares.17,18

December 2010

The present analysis examined flare rates in the 3

Phase III trials of febuxostat, based on mean postbaseline sUA concentrations and the duration of prophylaxis. AEs were assessed by prophylaxis with colchicine or naproxen.

PATIENTS AND METHODS

This was an investigator-initiated, post hoc reanalysis
of data on gout flares from the 3 Phase III randomized
controlled trials of febuxostatFACT (Febuxostat
Versus Allopurinol Control Trial in Subjects With
Gout) (NCT00102440),9 APEX (Allopurinol- and
Placebo-Controlled, Efficacy Study of Febuxostat)
(NCT00174915),10 and CONFIRMS (A Phase 3, Randomized, Multicenter, Double-Blind, Allopurinol-Controlled Study Assessing the Efficacy and Safety of Oral
Febuxostat in Subjects With Gout) (NCT00430248).11
Patients in the Phase III trials were males or females
aged 18 85 years who had a diagnosis of gout (American Rheumatology Association preliminary criteria19)
and a baseline sUA concentration 8.0 mg/dL. The
studies included patients with mild or moderate renal
impairment (estimated creatinine clearance [eCrCl]
60 89 and 30 59 mL/min, respectively, as determined
by the Cockroft-Gault formula corrected for ideal
body weight).20,21 Exclusion criteria included secondary hyperuricemia; inability to tolerate allopurinol,
naproxen, or colchicine; and severe renal impairment
(eCrCl 30 mL/min).
Table I summarizes the main details of the trial designs. Patients in the 3 studies received ULT with febuxostat, allopurinol, or placebo (6 months or 1 year) and
flare prophylaxis with either colchicine 0.6 mg/d or
naproxen 250 mg BID (8 weeks or 6 months). The prophylactic regimen was chosen at the discretion of the investigator, based on renal function and known intolerance to either drug. Patients with an eCrCl 50 mL/min
were not to receive naproxen. The protocols of all 3 studies received independent review board approval. The
sponsor of the trials (Takeda Global Research and Development Center, Inc) provided all relevant data to the authors of the present analysis.
In the present analysis, the proportion of patients
requiring treatment for gout flares during each trial
was reported by mean postbaseline sUA 6.0 or
6.0 mg/dL. Rates for each trial were reported at
4-week intervals. AEs recorded during FACT and
APEX are combined because of their common duration of prophylaxis (8 weeks), and AEs recorded dur-

2387

Clinical Therapeutics

Table I. Designs of FACT (Febuxostat Versus Allopurinol Control Trial in Subjects With Gout),9 APEX (Allopurinol- and Placebo-Controlled, Efficacy Study of Febuxostat),10 and CONFIRMS (A Phase 3, Randomized,
Multicenter, Double-Blind, Allopurinol-Controlled Study Assessing the Efficacy and Safety of Oral
Febuxostat in Subjects With Gout).11
Variable
ULT regimens

FACT

APEX

Febuxostat 80 mg/d
Febuxostat 120 mg/d
Febuxostat 240 mg/d
Allopurinol 300/100 mg/d*
Placebo
Randomization 1:1:1
2:2:1:2:1
Duration
2-Wk washout in those receiving 2-Wk washout in those
previous ULT; 52 wk of study
receiving previous ULT; 28
drug after randomization
wk of study drug after
randomization
Flare
Colchicine 0.6 mg/d or naproxen Colchicine 0.6 mg/d or
prophylaxis
250 mg BID during washout
naproxen 250 mg BID
period and for first 8 wk of
during washout period and
study; prophylaxis administered
for first 8 wk of study;
at discretion of investigator;
prophylaxis administered at
colchicine recommended for
discretion of investigator;
patients with sCr 1.5 mg/dL
colchicine recommended
for patients with sCr 1.5
mg/dL

Study visits

CONFIRMS

Febuxostat 80 mg/d
Febuxostat 120 mg/d
Allopurinol 300 mg/d

Febuxostat 40 mg/d
Febuxostat 80 mg/d
Allopurinol 300/200 mg/d*

Wk 2 and 4, monthly thereafter

1:1:1
30-Day washout in those
receiving previous ULT; 6
mo of study drug after
randomization
Colchicine 0.6 mg/d or
naproxen 250 mg BID
during washout period and
for entire 6 mo of study;
patients who received
naproxen were also
administered lansoprazole
15 mg/d; choice of
prophylaxis made by
investigator and patient,
taking into account
previous tolerance;
colchicine recommended
for patients with eCrCl 50
mL/min
Mo 2, 4, and 6

Wk 2, 4, 6, and 8, monthly
thereafter

eCrCl estimated creatinine clearance; sCr serum creatinine; ULT urate-lowering therapy.
* In APEX, patients who were randomized to receive allopurinol and had an sCr 1.5 mg/dL received a dose of 100 mg/d.
Similarly, in CONFIRMS, patients who were randomized to receive allopurinol and had an eCrCl of 30 59 mL/min (moderate
renal impairment) received a dose of 200 mg/d.

At each visit, sUA was measured in a blinded fashion, laboratory assessments were performed, and gout flares, adverse events,
and concomitant medication use were recorded.

ing CONFIRMS are reported separately. AEs were

summarized by receipt of prophylaxis with colchicine
or naproxen. All patients who received at least one
dose of study drug and at least one dose of prophylactic
medication were included in the tolerability analysis.

2388

RESULTS
Demographic and Baseline Characteristics
Across the Phase III studies, the majority of patients were white (80.1%), male (94.5%), and obese
(body mass index 30 kg/m2) (62.8%) (Table II).

Volume 32 Number 14

R.L. Wortmann et al.

Table II. Demographic and baseline characteristics of patients in the Phase III studies. Data are number (%) of
patients, unless otherwise specified.
Characteristic

FACT9 (N 760)

APEX10 (N 1072)

CONFIRMS11 (N 2269)

729 (95.9)

1005 (93.8)

2141 (94.4)

31 (4.1)

67 (6.3)

128 (5.6)

Sex*
Male
Female
Race
587 (77.2)

835 (77.9)

1863 (82.1)

Black/African American

White

62 (8.2)

120 (11.2)

228 (10.0)

Asian

25 (3.3)

26 (2.4)

88 (3.9)

Native Hawaiian/Pacific Islander

18 (2.4)

20 (1.9)

32 (1.4)

Native American/Alaskan Native

1 (0.1)

1 (0.1)

22 (1.0)

67 (8.8)

70 (6.5)

34 (1.5)

2 (0.1)

51.8 (12.13)

51.5 (12.17)

52.8 (11.73)

2383

2284

1985

45 y

230 (30.3)

312 (29.1)

568 (25.0)

4565 y

398 (52.4)

597 (55.7)

1327 (58.5)

65 y

132 (17.4)

163 (15.2)

374 (16.5)

11.9 (9.56)

10.9 (8.96)

11.6 (9.31)

151

141

153

9.8 (1.24)

9.9 (1.26)

9.6 (1.18)

9.0 mg/dL

207 (27.2)

285 (26.6)

837 (36.9)

9.010.0 mg/dL

236 (31.1)

366 (34.1)

699 (30.8)

10.011.0 mg/dL

181 (23.8)

221 (20.6)

454 (20.0)

11.012.0 mg/dL

87 (11.4)

125 (11.7)

189 (8.3)

12.0 mg/dL

45 (5.9)

70 (6.5)

90 (4.0)

186 (24.5)

299 (27.9)

478 (21.1)

Cardiovascular disease

74 (9.7)

144 (13.4)

1297 (57.2)

Diabetes

53 (7.0)

90 (8.4)

312 (13.8)

Hyperlipidemia

258 (33.9)

352 (32.8)

942 (41.5)

Hypertension

333 (43.8)

508 (47.4)

1199 (52.8)

Other
Missing data
Age, y
Mean (SD)
Range
Age group*

Duration of gout, y
Mean (SD)
Range
Serum urate, mean (SD), mg/dL
Serum urate category

History of tophi
Comorbidities

Body mass index 30 kg/m2

472 (62.1)

662 (61.8)

1442 (63.6)

Alcohol drinker

502 (66.1)

709 (66.1)

1549 (68.3)

Tobacco user

131 (17.2)

216 (20.1)

410 (18.1)

Creatinine clearance category*

90 mL/min (normal)

371 (48.8)

541 (50.5)

786 (34.6)

6089 mL/min (mild impairment)

295 (38.8)

377 (35.2)

1081 (47.6)

94 (12.4)

154 (14.4)

402 (17.7)

3059 mL/min (moderate impairment)

APEX Allopurinol- and Placebo-Controlled, Efficacy Study of Febuxostat; CONFIRMS A Phase 3, Randomized, Multicenter, Double-Blind, Allopurinol-Controlled Study Assessing the Efficacy and Safety of Oral Febuxostat in Subjects With Gout;
FACT Febuxostat Versus Allopurinol Control Trial in Subjects With Gout.
* Percentages may not total 100 due to rounding.

Serum urate concentrations in FACT and APEX were reported for the intent-to-treat cohort; therefore, within this category, the
patient numbers for each of these trials differ from the numbers of enrolled patients.

Patients with hypertension were included in the count of patients with cardiovascular disease in CONFIRMS, but not in FACT and APEX.

Values are estimated creatinine clearance, calculated using the Cockroft-Gault formula corrected for ideal body weight.20,21

December 2010

2389

Clinical Therapeutics
The mean duration of gout ranged from 10.9 11.9
years, and the mean baseline sUA concentration ranged from 9.6 9.9 mg/dL. Of the 760 patients
in FACT, 416 (54.7%) received gout flare prophylaxis with colchicine and 339 (44.6%) received prophylaxis with naproxen. Of the 1072 patients
in APEX, 577 (53.8%) received prophylaxis with
colchicine and 490 (45.7%) received prophylaxis
with naproxen. Five patients in each of these
2 studies (1% each) received no prophylaxis. Of
the 2269 patients in CONFIRMS, 1807 (79.6%),
346 (15.2%), and 116 (5.1%) received colchicine,
naproxen, or another form of prophylaxis (eg, celecoxib, indomethacin, nabumetone, or prednisone),
respectively.

Patients Requiring Treatment for Flares

The proportions of patients requiring treatment
for gout flares in the Phase III trials, summarized by
mean postbaseline sUA 6.0 or 6.0 mg/dL, are
illustrated in Figure 1. In the FACT and APEX trials,
in which prophylaxis was administered for 8 weeks,
the rate of flares increased markedly after the cessation of prophylaxis and then declined steadily. In
contrast, in the CONFIRMS trial, during which prophylaxis was administered for all 6 months of the
study, flare rates declined steadily following the initiation of ULT. Because patients were not followed
after the completion of CONFIRMS, the frequency

of flare rates after the cessation of both prophylaxis

and ULT is not known.
Flare rates were comparable in patients with sUA concentrations 6.0 and 6.0 mg/dL through weeks 1216
of FACT. Among patients with a mean postbaseline sUA
concentration 6.0 mg/dL, flare rates during the first 4
weeks and weeks 4 8, 8 12, and 1216 were 16%,
16%, 36%, and 28%, respectively. The corresponding
values in patients with a mean postbaseline sUA concentration 6.0 mg/dL were 18%, 13%, 37%, and 27%.
However, by the final 4 weeks, the mean rate of flares
among patients with a mean postbaseline sUA concentration 6.0 mg/dL was significantly lower than that among
patients with a mean postbaseline sUA concentration
6.0 mg/dL (6% vs 14%, respectively; P 0.05) (Figure
1A). During the first 12 weeks of APEX, patients with a
mean postbaseline sUA concentration 6.0 mg/dL had
numerically higher flare rates compared with those with a
mean postbaseline sUA concentration 6.0 mg/dL
(range, 14%26%); however, this pattern was reversed
after weeks 12 and 16 and, by the end of the study, patients with a mean postbaseline sUA concentration 6.0
mg/dL subsequently had numerically lower flare rates
compared with those with a mean postbaseline sUA concentration 6.0 mg/dL (13% and 18%, respectively)
(Figure 1B). During CONFIRMS, patients with a mean
postbaseline sUA concentration 6.0 mg/dL had consistently numerically lower flare rates beginning with
10% during the first 4 weeks of the study and steadily

Figure 1 A. Patients requiring treatment for flares, by mean postbaseline serum urate (sUA) concentration, in
FACT (Febuxostat Versus Allopurinol Control Trial in Subjects With Gout).
(continued)

2390

Volume 32 Number 14

R.L. Wortmann et al.

Figure 1 B. (continued). Patients requiring treatment for flares, by mean postbaseline serum urate (sUA) concentration, in APEX (Allopurinoland Placebo-Controlled, Efficacy Study of Febuxostat).10

declining to 3% during the last 4 weeks compared with

patients with a mean postbaseline sUA concentration
6.0 mg/dL, in whom flare rates during the first 4 weeks
were 11% and declined to 5% by weeks 24 28 (Figure
1C).

Figure 2 illustrates the reduction in flare rates in the

last 4 weeks (weeks 48 52) of FACT after 1 year of
ULT, summarized by mean postbaseline sUA concentration. Patients with lower mean postbaseline sUA
concentrations had numerically fewer flares compared

Figure 1 C. (continued). Patients requiring treatment for flares, by mean postbaseline serum urate (sUA)
concentration, in CONFIRMS (A Phase 3, Randomized, Multicenter, Double-Blind, AllopurinolControlled Study Assessing the Efficacy and Safety of Oral Febuxostat in Subjects With Gout).11
In CONFIRMS, prophylaxis was administered throughout the entire study period. Reproduced with
permission.3

December 2010

2391

Clinical Therapeutics

Figure 2. Proportion of patients requiring treatment for gout flares at the end of 1 year of treatment, by mean
postbaseline serum urate (sUA) concentration, in FACT (Febuxostat Versus Allopurinol Control Trial
in Subjects With Gout).9 *P 0.05.

with those with higher mean postbaseline sUA concentrations. Specifically, flare rates among patients with
mean postbaseline sUA concentrations 8.0 and 7.0 to
8.0 mg/dL were 18.2% and 16.3%, respectively,
whereas rates among patients with mean postbaseline
sUA concentrations of 4.0 to 5.0 and 4.0 mg/dL were
5.3% and 4.5%, respectively.
The positive effect of prophylaxis for 6 months
(CONFIRMS) relative to 8 weeks (FACT and APEX)
was reflected in rates of premature discontinuation owing to flares. Of patients who discontinued the trials
prematurely, 18.7% (47/252), 9.3% (28/300), and
2.9% (12/418) of patients did so because of gout flares
in FACT, APEX, and CONFIRMS, respectively.

Tolerability
Pooled rates of overall AEs in FACT and APEX
were significantly higher among patients who received
prophylaxis with colchicine compared with naproxen
(55.1% vs 44.3%, respectively; P 0.001) (Table III).
However, in CONFIRMS, rates of overall AEs did not
differ significantly between the 2 prophylaxis groups
(55.1% and 54.9%) (Table IV). In all 3 studies, upper
respiratory tract infection was the most frequently reported AE (occurring in 2% of patients); the fre-

2392

quency of this AE did not differ significantly between

colchicine and naproxen.
There were, however, differences in rates of individual AEs. During FACT and APEX, in which prophylaxis was given for only the first 8 weeks, diarrhea was
reported more than 3 times as often in patients receiving colchicine compared with those receiving naproxen
(8.4% vs 2.7%, respectively; P 0.001) (Table III).
During CONFIRMS, in which prophylaxis was given
for the entire 6-month period, there was no significant
difference in the incidence of diarrhea between colchicine and naproxen (5.1% and 6.4%) (Table IV). In
CONFIRMS, those who received naproxen had significantly higher rates of gastrointestinal and abdominal
pain compared with those who received colchicine
(3.2% vs 1.2%, respectively; P 0.012), as well as
significantly higher rates of dental and oral soft tissue
infections (2.3% vs 0.6%; P 0.006, respectively).
Conversely, rates of headaches were 3 times higher in
the colchicine group compared with the naproxen
group (2.8% vs 0.9%, respectively; P 0.037).
In CONFIRMS, investigator-reported elevations in
measures of liver function were significantly more frequent among patients receiving flare prophylaxis with
colchicine than with naproxen (7.7% vs 4.3%, respec-

Volume 32 Number 14

R.L. Wortmann et al.

Table III. Most frequently reported adverse events (AEs) (occurring in 2% of patients), by Medical Dictionary
for Regulatory Activities high-level term, among patients receiving colchicine or naproxen as flare
prophylaxis in FACT (Febuxostat Versus Allopurinol Control Trial in Subjects With Gout)9 and APEX
(Allopurinol- and Placebo-Controlled, Efficacy Study of Febuxostat).10 Values are number (%) of
patients.
Variable
Total no. of patients reporting 1 AE
Most frequent AEs
Upper respiratory tract infections
Diarrhea
Musculoskeletal and connective tissue signs and symptoms
Joint-related signs and symptoms
Headaches
Nausea and vomiting symptoms
Liver function analyses
Neurologic signs and symptoms
Nonsite-specific injuries
Gastrointestinal and abdominal pains

Colchicine (n 993)

Naproxen (n 829)

547 (55.1)*

367 (44.3)

90 (9.1)
83 (8.4)*
57 (5.7)
52 (5.2)
49 (4.9)
35 (3.5)
25 (2.5)
25 (2.5)
23 (2.3)
20 (2.0)

63 (7.6)
22 (2.7)
34 (4.1)
30 (3.6)
34 (4.1)
23 (2.8)
16 (1.9)
12 (1.4)
8 (1.0)
17 (2.1)

* P 0.001, Fisher exact test.

P 0.029, Fisher exact test.

tively; P 0.023). Alanine aminotransferase values

2 times the upper limit of normal (ULN) occurred
significantly more frequently in patients receiving colchicine than in those receiving naproxen (10% vs 5%;
P 0.014). The percentage of patients with aspartate
aminotransferase values 2 times the ULN did not
differ significantly between prophylaxis groups. There
were no cases of abnormal or elevated liver function
test results that satisfied Hys rule, defined as the occurrence of alanine aminotransferase or aspartate aminotransferase values 3 times the ULN, along with
elevations in bilirubin 2 times the ULN that could not
be explained by viral hepatitis, preexisting or acute
liver disease, or another drug capable of causing liver
injury.22

DISCUSSION
Hyperuricemia (sUA 6.8 mg/dL), the underlying
pathophysiology in gout, leads to the supersaturation
of body fluids with urate. Under permissive conditions,
such as decreased temperature or lack of protein binding, urate crystal formation and deposition may occur.23 However, the majority of hyperuricemic individuals never experience a gout flare.24 26 The reasons for

December 2010

this are not clear. Several additional clinical risk factors

for the development of gout have been identified, including male sex, higher body mass index, greater age,
hypertension, and chronic renal failure.24,25
Although increases in urate concentrations may
provoke a flare, it is also common for acute gouty arthritis to occur when urate concentrations drop. This
may coincide with the initiation of ULT with xanthine
oxidase inhibitors or uricosuric agents. Many earlier
studies of ULT reported this paradox,4,6 8,27 and the
data from FACT,9 APEX,10 and CONFIRMS11 suggest that gout flares frequently do follow the initiation
of ULT.
Post hoc analyses of flare data from FACT and
APEX revealed that a greater relative reduction in sUA
was associated with a persistent increase in the risk for
flares.3 In addition, a lower absolute postbaseline sUA
concentration early in ULT is associated with an increased risk for flares, supporting the hypothesis that
rapid decreases in sUA are likely to trigger flares. However, this effect ceases to be observed as the duration of
ULT increases.3 In a study in 339 patients with acute
gout flares, 16 of 55 (29.1%) patients who received
allopurinol and had a baseline sUA concentration

2393

Clinical Therapeutics

Table IV. Most frequently reported adverse events (AEs) (occurring in 2% of patients), by Medical Dictionary
for Regulatory Activities high-level term, among patients receiving colchicine or naproxen as flare
prophylaxis in CONFIRMS (A Phase 3, Randomized, Multicenter, Double-Blind, Allopurinol-Controlled Study Assessing the Efficacy and Safety of Oral Febuxostat in Subjects With Gout).11 Values are
number (%) of patients.
Variable
Total no. of patients reporting 1 AE
Most frequent AEs
Upper respiratory tract infections
Liver function analyses
Musculoskeletal and connective tissue signs and symptoms
Diarrhea
Joint-related signs and symptoms
Headaches
Dermatitis and eczema
Nausea and vomiting symptoms
Vascular hypertensive disorders
Lower respiratory tract and lung infections
Rashes, eruptions, and exanthems
Edema
Gastrointestinal and abdominal pains
Dental and oral soft tissue infections

Colchicine (n 1807)

Naproxen* (n 346)

996 (55.1)

190 (54.9)

150 (8.3)
140 (7.7)
94 (5.2)
93 (5.1)
66 (3.7)
51 (2.8)
47 (2.6)
40 (2.2)
37 (2.0)
31 (1.7)
28 (1.5)
26 (1.4)
21 (1.2)
11 (0.6)

27 (7.8)
15 (4.3)
14 (4.0)
22 (6.4)
17 (4.9)
3 (0.9)
4 (1.2)
6 (1.7)
11 (3.2)
8 (2.3)
8 (2.3)
7 (2.0)
11 (3.2)
8 (2.3)

* Patients receiving naproxen for flare prophylaxis also received lansoprazole 15 mg/d.

P 0.023, Fisher exact test.

P 0.037, Fisher exact test.

P 0.012, Fisher exact test.

P 0.006, Fisher exact test.

6.0 mg/dL experienced flares over a 7-day period.28

Continued flares in patients receiving allopurinol were
attributed to the persistence of tophi, increased body
uric acid pools, and the need for larger doses or a longer duration of allopurinol treatment. In fact, longterm ULT and maintenance of an sUA concentration
6.0 mg/dL for at least 1 year have been associated
with a significantly reduced incidence of acute gout.
When flare rates in CONFIRMS11 were summarized
by previous participation in either of the open-label,
long-term extension studies (ULT for up to 5
years),17,18 the rate of flares was significantly lower in
patients with previous participation compared with
those without previous participation (P 0.001). It
should be noted that patients were not followed after
completing CONFIRMS; therefore, the frequency of
flare rates after the cessation of both prophylaxis and
ULT is not known. However, in a prospective, long-

2394

term follow-up study (mean duration of follow-up,

26.4 months) in 89 patients who had been receiving
ULT for at least 5 years (mean duration of ULT, 89
months) and then stopped ULT, patients in whom the
sUA concentration had been maintained below the
subsaturating range (5.05 mg/dL) throughout ULT
were significantly more likely to have a prolonged period without gout symptoms after the withdrawal of
ULT compared with those whose sUA concentration
remained 5.05 mg/dL (mean survival time, 48.8 vs
33.9 months, respectively; P 0.01).29
By weeks 48 52 of FACT, flare rates were lowest
among patients with the lowest postbaseline sUA
concentration. Furthermore, whereas 1 year of ULT
in FACT was associated with the need to treat flares
in 15% of patients overall, continued use of ULT
for 3 to 5 years was associated with almost complete
elimination of flares.17,18

Volume 32 Number 14

R.L. Wortmann et al.

Rapid lowering of sUA may induce activation of
previously precipitated urate crystals in cartilage and
soft tissues. These precipitates trigger a flare by stimulating cyclooxygenase-2 expression and prostaglandin
E2 production by monocytes.30,31 The phagocytic capacity of neutrophils may also be enhanced as urate
levels decrease.32 Finally, if urate concentrations decrease to 6.8 mg/dL, superficial crystals may become
solubilized, exposing uncoated crystals to monocytes
and synoviocytes. These crystals activate Toll-like receptors 2 and 4 and engage the NALP-3 (NACHT,
LRR, and pyrin domain containing protein3) inflammasome, causing expression of numerous nuclear factor-regulated inflammatory cytokines and resulting in production of active interleukin (IL)-1 and
IL-18.33,34
The European League Against Rheumatism guidelines for gout recommend that prophylaxis be administered during the early months of ULT to reduce the
risk of flares.1 Low-dose colchicine has long been considered the mainstay of prophylaxis for acute gout
flare, and clinical studies have reported significant reductions in flares when colchicine was administered in
conjunction with allopurinol (P 0.008)13 or probenecid (P 0.05)14 compared with either ULT alone.
NSAIDs may also be used to reduce the risk of flares
during ULT.1 Although the studies reported here did
not have a comparator arm in which patients did not
receive flare prophylaxis, the low rate of discontinuations due to flares in CONFIRMS (2.9%) compared
with FACT and APEX (18.7% and 9.3%, respectively)
suggest that continuing prophylaxis for 6 months may
avoid the increased risk of flaring seen when prophylaxis was withdrawn after 8 weeks of ULT. The increase in flare rates after the cessation of prophylaxis
may reflect the continued mobilization of urate from
body pools.
Previous recommendations for the duration of prophylaxis have ranged from 1 month to 1 year.35,36 The
results of the present analysis support continuation of
prophylaxis for up to 6 months. Clinical data are not
yet available to clarify whether withdrawing prophylaxis after 6 months while continuing ULT is likely to
be followed by a spike in flares; by 6 months, patients
responsive to ULT have already achieved and maintained an sUA concentration 6.0 mg/dL. Therefore,
there is reduced mobilization of urate from body pools
at this time and a lower chance of inflammasome activation. A correlation has been found between disease

December 2010

duration and the duration of ULT needed to eliminate

urate crystals,37 suggesting that patients with a longer
duration of disease may require a longer period of flare
prophylaxis.
Both colchicine 0.6 mg/d and naproxen 250 mg BID
were well tolerated in the population studied. The most
common AEs included upper respiratory infections,
musculoskeletal and connective tissue problems, and
diarrhea. Diarrhea was significantly more frequent
with colchicine than with naproxen when prophylaxis
was given for only 8 weeks (P 0.001); on the other
hand, diarrhea was numerically more frequent with
naproxen than with colchicine when prophylaxis was
continued for 6 months. During CONFIRMS, patients
receiving naproxen were also administered the proton
pump inhibitor lansoprazole (15 mg/d). Diarrhea is the
most commonly reported AE with lansoprazole,38 possibly explaining the increased rates of this AE in the
present analysis. Liver function abnormalities were
more frequent with colchicine than with naproxen in
CONFIRMS.
The data presented here are limited by the post hoc
nature of the analyses. Prospective randomized controlled trials are needed to evaluate flare prophylaxis
for up to 6 months, after 6 months, and over time.
These results are consistent with the findings of earlier
randomized controlled trials, although the doses of
colchicine used in those studies were higher than those
used in the febuxostat Phase III trials, and the extent of
urate lowering, which directly affects flare rates, was
not reported.13,14

CONCLUSIONS
The results of this post hoc analysis suggest that prophylaxis with low-dose colchicine or naproxen reduced the risk of gout flares after the initiation of ULT.
Based on the difference in flare rates in CONFIRMS
compared with FACT and APEX, 6 months of prophylaxis may impart a greater benefit than 2 months, with
no associated increase in AEs.

ACKNOWLEDGMENTS
FACT, APEX, and CONFIRMS were funded by TAP
Pharmaceutical Products, Inc, now a part of Takeda
Global Research and Development Center, Inc. This
was an investigator-initiated reanalysis. Dr. Wortmann has served as a speaker and consultant for
Takeda Global Research and Development Center. He
was provided with all data for the reanalysis and was

2395

Clinical Therapeutics
not paid for being an author. His authorship is in full
compliance with Dartmouth Medical School academic
guidelines for authorship, and all authors met International Committee of Medical Journal Editors requirements for authorship. Ms. MacDonald, Ms. Hunt, and
Dr. Jackson are employees of Takeda Global Research
and Development Center. None of the Takeda employees receive compensation in relation to the success of
the study product. The authors have indicated that
they have no other conflicts of interest with regard to
the content of this article.
The authors thank Solomon Chefo, PhD, Senior
Statistician, Takeda Global Research and Development Center, for critical review of the manuscript. Dr.
Chefo and Ms. Hunt performed the statistical analyses.
Assistance in manuscript preparation was provided by
Meryl Gersh, PhD, of AlphaBioCom, LLC, in Radnor,
Pennsylvania, and was funded by Takeda Global Research and Development Center.

REFERENCES
1. Zhang W, Doherty M, Bardin T, et al; for the EULAR
Standing Committee for International Clinical Studies
Including Therapeutics. EULAR evidence based recommendations for gout. Part II: Management. Report of a task
force of the EULAR Standing Committee for International
Clinical Studies Including Therapeutics (ESCISIT). Ann
Rheum Dis. 2006;65:13121324.
2. Shoji A, Yamanaka H, Kamatani N. A retrospective study of
the relationship between serum urate level and recurrent
attacks of gouty arthritis: Evidence for reduction of recurrent gouty arthritis with antihyperuricemic therapy. Arthritis
Rheum. 2004;51:321325.
3. Becker MA, MacDonald PA, Hunt BJ, et al. Determinants of
the clinical outcomes of gout during the first year of
urate-lowering therapy. Nucleosides Nucleotides Nucleic Acids.
2008;27:585591.
4. Houpt JB. The effect of allopurinol (HPP) in the treatment
of gout. Arthritis Rheum. 1965;8:899 904.
5. Ogryzlo MA, Harrison J. Evaluation of uricosuric agents in
chronic gout. Ann Rheum Dis. 1957;16:425 437.
6. Wilson JD, Simmonds HA, North JD. Allopurinol in the
treatment of uraemic patients with gout. Ann Rheum Dis.
1967;26:136 142.
7. Yu TF, Burns JJ, Gutman AB. Results of a clinical trial of
G-28315, a sulfoxide analog of phenylbutazone, as a
uricosuric agent in gouty subjects. Arthritis Rheum. 1958;1:
532543.
8. Yue TF, Gutman AB. Effect of allopurinol (4-hydroxypyrazolo-(3,4-D)pyrimidine) on serum and urinary uric acid in
primary and secondary gout. Am J Med. 1964;37:885 898.

2396

9. Becker MA, Schumacher HR Jr, Wortmann RL, et al.

Febuxostat compared with allopurinol in patients with hyperuricemia and gout. N Engl J Med. 2005;353:2450 2461.
10. Schumacher HR Jr, Becker MA, Wortmann RL, et al. Effects
of febuxostat versus allopurinol and placebo in reducing
serum urate in subjects with hyperuricemia and gout: A
28-week, phase III, randomized, double-blind, parallelgroup trial. Arthritis Rheum. 2008;59:1540 1548.
11. Becker MA, Schumacher HR, Espinoza LR, et al. The
urate-lowering efficacy and safety of febuxostat in the
treatment of the hyperuricemia of gout: The CONFIRMS
trial. Arthritis Res Ther. 2010;12:R63.
12. Schlesinger N. Management of acute and chronic gouty
arthritis: Present state-of-the-art. Drugs. 2004;64:
2399 2416.
13. Borstad GC, Bryant LR, Abel MP, et al. Colchicine for
prophylaxis of acute flares when initiating allopurinol for
chronic gouty arthritis. J Rheumatol. 2004;31:2429 2432.
14. Paulus HE, Schlosstein LH, Godfrey RG, et al. Prophylactic
colchicine therapy of intercritical gout. A placebo-controlled study of probenecid-treated patients. Arthritis Rheum.
1974;17:609 614.
15. Takano Y, Hase-Aoki K, Horiuchi H, et al. Selectivity of
febuxostat, a novel non-purine inhibitor of xanthine oxidase/
xanthine dehydrogenase. Life Sci. 2005;76:18351847.
16. Uloric [package insert]. Deerfield, Ill: Takeda Pharmaceuticals America, Inc; 2009.
17. Becker MA, Schumacher HR, MacDonald PA, et al. Clinical
efficacy and safety of successful longterm urate lowering
with febuxostat or allopurinol in subjects with gout.
J Rheumatol. 2009;36:12731282.
18. Schumacher HR Jr, Becker MA, Lloyd E, et al. Febuxostat in
the treatment of gout: 5-Yr findings of the FOCUS efficacy
and safety study. Rheumatology (Oxford). 2009;48:188 194.
19. Wallace SL, Robinson H, Masi AT, et al. Preliminary criteria
for the classification of the acute arthritis of primary gout.
Arthritis Rheum. 1977;20:895900.
20. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976;16:31 41.
21. Robert S, Zarowitz BJ, Peterson EL, Dumler F. Predictability
of creatinine clearance estimates in critically ill patients.
Crit Care Med. 1993;21:14871495.
22. US Food and Drug Administration. Guidance for industry.
Drug-induced liver injury: Premarketing clinical evaluation.
http://www.fda.gov/downloads/Drugs/GuidanceCompliance
RegulatoryInformation/Guidances/UCM174090.pdf. Accessed September 22, 2010.
23. Kippen I, Klinenberg JR, Weinberger A, Wilcox WR. Factors
affecting urate solubility in vitro. Ann Rheum Dis. 1974;33:
313317.
24. Campion EW, Glynn RJ, DeLabry LO. Asymptomatic
hyperuricemia. Risks and consequences in the Normative
Aging Study. Am J Med. 1987;82:421 426.

Volume 32 Number 14

R.L. Wortmann et al.

25. Zhang W, Doherty M, Pascual E, et
al; for EULAR Standing Committee
for International Clinical Studies Including Therapeutics. EULAR evidence based recommendations for
gout. Part I: Diagnosis. Report of a
task force of the Standing Committee for International Clinical Studies
Including Therapeutics (ESCISIT).
Ann Rheum Dis. 2006;65:13011311.
26. Luk AJ, Simkin PA. Epidemiology of
hyperuricemia and gout. Am J Manag
Care. 2005;11(Suppl):S435S442.
27. Kuzell W, Glover R, Gibbs J, et al.
Effect of anturane on serum uric
acid and cholesterol in gout: A longterm study. Acta Rheumatol Scand
Suppl. 1964;8:31 40.
28. Schlesinger N, Norquist JM, Watson
DJ. Serum urate during acute gout.
J Rheumatol. 2009;36:12871289.
29. Perez-Ruiz F, Atxotegi J, Hernando I,
et al. Using serum urate levels to
determine the period free of gouty
symptoms after withdrawal of longterm urate-lowering therapy: A prospective study. Arthritis Rheum. 2006;
55:786 790.
30. Fam AG. Treating acute gouty arthritis with selective COX 2 inhibitors.
BMJ. 2002;325:980 981.
31. Pouliot M, James MJ, McColl SR, et
al. Monosodium urate microcrystals induce cyclooxygenase-2 in human monocytes. Blood. 1998;91:
1769 1776.
32. Li EK. Gout: A review of its aetiology
and treatment. Hong Kong Med J.
2004;10:261270.
33. Liu-Bryan R, Scott P, Sydlaske A, et
al. Innate immunity conferred by
Toll-like receptors 2 and 4 and myeloid differentiation factor 88 expression is pivotal to monosodium urate
monohydrate crystal-induced inflammation. Arthritis Rheum. 2005;
52:2936 2946.
34. Martinon F, Ptrilli V, Mayor A, et al.
Gout-associated uric acid crystals
activate the NALP3 inflammasome.
Nature. 2006;440:237241.
35. Emmerson BT. The management of
gout. N Engl J Med. 1996;334:445451.

December 2010

36. Wortmann RL. Effective management of gout: An analogy. Am J Med.

1998;105:513514.
37. Pascual E, Sivera F. Time required
for disappearance of urate crystals
from synovial fluid after successful

hypouricaemic treatment relates to

the duration of gout. Ann Rheum Dis.
2007;66:1056 1058.
38. Prevacid [package insert]. Deerfield,
Ill: Takeda Pharmaceuticals America, Inc; 2008.

Address correspondence to: Robert L. Wortmann, MD, Section of Rheumatology, Dartmouth Medical School, One Medical Center Drive, Lebanon, NH 03756. E-mail: Robert.L.Wortmann@Hitchcock.org

2397

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.