Use of febuxostat in the management of gout in the United Kingdom

Abstract

Gout is the most common cause of inflammatory arthritis worldwide. Despite clinical cure being achievable and multiple evidence-based guidelines having been published, the incidence and prevalence continues to increase and the condition remains undertreated. Concerns regarding allopurinol have limited its use in those with renal impairment. Febuxostat, a novel xanthine oxidase inhibitor requiring no dose adjustment in mild−moderate renal impairment was launched in the United Kingdom (UK) in 2010. We review published data on the efficacy, safety and tolerability of febuxostat and provide an opinion on its place in the management of gout in the UK in the context of other published guidelines. One phase II trial, multiple phase III trials [febuxostat versus allopurinol controlled trial (FACT), APEX, CONFIRMS] and two open-label extension trials have demonstrated febuxostat given at the doses commonly used in UK practice (80 mg, 120 mg) to reduce serum urate more effectively than those receiving fixed-dose allopurinol. Overall adverse event rates were comparable across treatment groups aside from gout flare (more common in febuxostat-treated patients) and concerns regarding cardiovascular toxicity are being further evaluated in two large trials. If the outcomes of these are favourable, we would anticipate a marked increase in the use of febuxostat in the UK market. We would advocate the use of febuxostat to target a serum urate < 0.3 mmol/l (5 mg/dl) as a second-line urate-lowering therapy in patients with hyperuricaemia, and clinical gout in those intolerant of allopurinol, or in those in whose renal function precludes optimal dose escalation to achieve target serum urate. We would advise prophylaxis against gouty flare with colchicine, nonsteroidal anti-inflammatory drugs (NSAIDs), or Cyclo-oxygenase-2 selective NSAID (COXIB) after febuxostat initiation.

Epidemiology and scope of the problem

Gout is the most common cause of inflammatory arthritis worldwide [Roddy and Doherty, 2010]. In the United Kingdom (UK), epidemiological studies in general practice have shown the prevalence of gout per 1000 to have been steadily increasing from 2.6 in 1975 [Currie, 1979], 3.4 in 1987 [Steven, 1992], to 9.5 in 1993 [Harris et al. 1995]. More recent work has shown that incidence and prevalence were both significantly higher in 2012 than 1997, with a 63.9% increase in prevalence and 29.6% increase in incidence over this period [Kuo et al. 2014]. It is found predominantly in middle-aged-to-elderly men [Roddy et al. 2013].

Despite the widespread availability of urate-lowering therapy at low cost for many years and the possibility of clinical ‘cure of gout’, studies show that the management of gout is often suboptimal. A large UK-based General Practice database [Kuo et al. 2014] showed that in 2012, urate-lowering therapy was only prescribed for one-quarter-to-one third of incident gout patients within a year of diagnosis, a figure unchanged since earlier estimates from 1990 to 1999 [Mikuls et al. 2005]. This is particularly disappointing, given the publishing of multiple evidence-based guidelines between these two studies’ time periods [Zhang et al. 2006; Jordan et al. 2007; Khanna et al. 2012].

Pathophysiology

The higher levels of serum urate seen in man as compared with other mammals is a consequence of our lacking hepatic urate oxidase, which functions to degrade uric acid, the end product of purine metabolism. The transformation of hypoxanthine to xanthine and then to uric acid is catalysed by xanthine oxidase.

As the serum urate level rises and the physiological saturation level of approximately 408 μmol/l at 37° [Fiddis et al. 1983] is exceeded, crystallization can occur with deposition into joints, soft tissues and the renal tract. This deposition may be clinically silent, with ultrasonography detecting tophi and inflammation in a third and in a quarter, respectively, of asymptomatic hyperuricaemic patients [Puig et al. 2008]. Aside from the most potent risk factor for gout of hyperuricaemia itself, other risk factors predicting the development of gout include hypertension, the use of loop and thiazide diuretics, obesity and high alcohol intake [Campion et al. 1987; Roubenoff et al. 1991; Waller and Ramsay, 1989]. Diets high in seafood, red meat or consumption of fructose-containing soft drinks are also established risk factors for incident gout, while dairy product consumption is protective [Choi et al. 2004a, 2004b; Choi and Curhan, 2008]. Additionally, recent genome-wide association studies have showed several genetic loci associated with hyperuricaemia and gout [Yang et al. 2010].

Asymptomatic uric acid deposition can progress to clinical acute gout when crystals are shed into synovial fluid. They are endocytosed by neutrophils and the NALP3 inflammasome [Martinon et al. 2006] is activated resulting in the release of IL-1beta into the extracellular environment. The resultant inflammation accounts for the symptoms and signs seen of intense pain, redness, heat and swelling that characterize acute gout attacks which most commonly affect peripheral joints, especially the first metatarsophalangeal joint [Wertheimer et al. 2013].

If serum urate remains above the saturation threshold, the patient is liable to experience further attacks of acute gout, which over time, may progress to deforming chronic gouty arthropathy [Becker and Ruoff, 2010] and tophi formation. renal failure due to gout is now uncommon.

Gout management

The mainstay of the management of acute gout is a nonsteroidal anti-inflammatory drug (NSAID) given at maximum dose, colchicine 500 mg given twice-to-four times a day or by intra-articular/intra-muscular or systemic corticosteroid [Jordan et al. 2007].

Once the acute attack has settled, consideration should be given to whether urate-lowering therapy should be prescribed, given that within 3 years, 80% of patients will have had a further attack of gout [Ferraz and O’Brien, 1995]. The British Society of Rheumatology (BSR) recommends that patients have a serum urate target of < 0.3 mmol/l (5 mg/dl) [Jordan et al. 2007].

Allopurinol

Although allopurinol remains the first-line urate-lowering therapy used in the UK, its use is limited in certain patients. Allopurinol is a pyrazolopyrimidine and analogue of hypoxanthine, working by reducing the oxidation of hypoxanthine to xanthine and xanthine to uric acid. It was, for many years, the only xanthine oxidase inhibitor available.

However, around 5% of patients cannot tolerate allopurinol and extremely rarely, severe and potentially life-threatening adverse events have occurred such as the allopurinol hypersensitivity syndrome (AHS), with a case fatality rate of up to 25% [Emmerson et al. 1988; Arellano and Sacristan, 1993]. Early work established that most cases of AHS occurred in those with renal impairment [Hande et al. 1984] and it has therefore been established best practice to adjust the allopurinol dose for renal function in efforts to reduce the risk of severe toxicity.

However, the exact role of renal impairment in the development of AHS has become more controversial over recent years. There is work suggesting that the starting dose of allopurinol, particularly in those with renal impairment, may predict the likelihood of severe adverse reaction to allopurinol rather than the ultimate dose of allopurinol given [Stamp et al. 2012]. More recently, a population Pharmacokinetic / pharmacodynamic (PKPD) model for allopurinol was developed to propose the maintenance doses of allopurinol that would be required to achieve target serum urate [Wright et al. 2016]. This tool showed that conventional dose reduction [Hande et al. 1984] for renal impairment would result in failure to achieve target urate, and performance at urate lowering would worsen with declining values of renal function. Interestingly, the authors found that the model predicted that allopurinol dose requirements were determined primarily by differences in body size and diuretic use, rather than by renal function, which had a modest impact on dosing, although this was more pronounced in those taking diuretics. We await with interest the results of research evaluating these predictions further, particularly with regard to safety of such high doses of allopurinol in the real-world gout population with significant renal impairment. Robust data on safety and efficacy would be needed before these changes would likely be adopted in clinical practice.

Thus, in patients with renal impairment, often the allopurinol dose has not been optimally uptitrated to achieve target serum urate concentration. Historically, these patients had few therapeutic options open to them aside from uricosurics with often limited effectiveness.

Febuxostat

Introduction

Febuxostat is a 2-arylthiazole derivative. It is a potent, nonpurine, selective inhibitor of both the oxidized and reduced forms of xanthine oxidase as compared with allopurinol, which only inhibits the reduced form. As xanthine oxidase is the catalyst of several steps in purine metabolism to uric acid, its inhibition results in reduced generation of uric acid. It is given orally, once daily and is rapidly and well absorbed. 99.2% of febuxostat is bound to plasma proteins. It is metabolized by conjugation via uridine diphosphate glucuronosyltransferase enzyme system and oxidation via the cytochrome P450 (CYP) system. It is eliminated via hepatic and renal pathways. Studies showed no dose adjustment to be necessary in mild−moderate renal or hepatic impairment or in elderly patients [European Medicines Agency, 2012].

Febuxostat is generally safe and well tolerated but the most commonly reported adverse reactions in trials and postmarketing surveillance were gout flares, liver function abnormalities, diarrhoea, nausea, headache, rash and oedema.

Guidance on the place of febuxostat in the management of gout in the UK

In the UK, febuxostat (Adenuric, Ipsen), was launched in 2010. The National Institute for Clinical Excellence (NICE) is a statutory body who perform technology appraisals to assess the clinical and cost effectiveness of health technologies, such as new pharmaceutical products to ensure that all National Health Service patients have equitable access to the most clinically and cost-effective treatments. In their Technology Appraisal Guideline of 2008, febuxostat was recommended for use in chronic hyperuricaemia in gout only for those intolerant of allopurinol (defined as adverse effects that are sufficiently severe to warrant its discontinuation, or to prevent full dose escalation for optimal effectiveness as appropriate within its marketing authorization) or for whom allopurinol is contraindicated [The National Institute for Health and Care Excellence, 2008].

The BSR is to publish updated guidelines for the management of gout in 2016 and these are expected to recommend the use of febuxostat as a second-line xanthine oxidase inhibitor to be used broadly in line with the NICE technology appraisal of 2008. The usual dosing regime is of 80 mg per day with dose escalation after 2–4 weeks to 120 mg per day to bring serum urate < 0.3 mmol/l (5 mg/dl).

Guidance on the place of febuxostat in the management of gout in the United States

The American College of Rheumatology, in their 2012 Guidelines for the Management of Gout [Khanna et al. 2012], do not recommend febuxostat over allopurinol or vice versa but allow the clinician’s discretion in selecting an appropriate urate-lowering therapy. In contrast to the BSR guidelines, it advises a variable serum urate target dependent on clinical manifestations of gout, suggesting a target of <0.36 mmol/l (6.0 mg/dl) in uncomplicated patients but <0.3 mmol/l (5 mg/dl) in those with greater disease severity (for example, those with tophi).

Efficacy of febuxostat

The efficacy of febuxostat as urate-lowering therapy in patients with hyperuricaemia and gout has now been investigated in a phase II trial (using the febuxostat doses usual in UK practice) and multiple phase III trials. There have also been two open-label extension trials providing data on longer-term efficacy and safety, and work is underway to look at the cardiovascular safety of the drug.

The first phase II randomized double-blind dose-response [Becker et al. 2005a] study was performed on 153 patients over 28 days who had a serum urate > 0.48 mmol/l (8.0 mg/dl) and a diagnosis of gout by the American College of Rheumatology Criteria (ACR) [Wallace et al. 1977]. Febuxostat 40 mg (n = 37), 80 mg (n = 40) and 120 mg (n = 38) per day were compared with placebo (n = 38), with a primary endpoint of the proportion of patients achieving a target serum urate of <0.36 mmol/l (6.0 mg/dl). Colchicine 600 mg twice a day (BD) was given as prophylaxis for the 2-week pretrial washout of any previously prescribed urate-lowering therapy and for the first 2 weeks of the trial. The primary endpoint was reached on day 28 by 0% of those taking placebo, in 56% of those taking 40 mg, 76% taking 80 mg, and 94% taking 120 mg of febuxostat. Flare of gout was common, particularly after the colchicine prophylaxis period ended and was more common with increasing doses of febuxostat. The drug was safe and well tolerated during this short trial. This trial was extended and 116 patients were enrolled in the FOCUS (Febuxostat Open Label of Urate-Lowering Efficacy and Safety) [Schumacher et al. 2009] which has now reported some 5-year efficacy and safety data (discussed later).

The febuxostat versus allopurinol controlled trial (FACT) [Becker et al. 2005b], published in 2005, was a phase III, randomized, double blind, 52-week, multicentre trial to compare the safety and efficacy of febuxostat with that of allopurinol in patients with gout and a serum urate > 0.48 mmol/l (8.0 mg/dl). It excluded those with a serum creatinine > 133 μmol/l. Patients were given NSAID or colchicine prophylaxis for the first 8 weeks of double-blind treatment. In the trial, 256 patients received 80 mg of febuxostat, 251 patients received 120 mg of febuxostat and 253 received 300 mg per day of allopurinol.

The primary endpoint in FACT was a serum urate of <0.36 mmol/l (<6.0 mg/dl) at the last 3-monthly measurements. The primary endpoint was reached in 53% of the 80 mg febuxostat group, by 62% of the 120 mg febuxostat group and by 21% of the allopurinol group. In the 156 subjects with tophi at baseline, a median percentage reduction of tophus area was seen of 83% of the 80 mg febuxostat group, 66% for the 120 mg febuxostat group and 50% for the allopurinol group.

The greater reduction in serum urate in the two febuxostat dosing arms was seen to be statistically significant by week 2 and was sustained throughout the study. It was therefore unsurprising that in the first 8 weeks, the incidence of gout flares was higher in the febuxostat 120 mg (36%) and febuxostat 80 mg (22%) groups compared to the allopurinol group (21%). Between weeks 9–52, a high proportion of the patients had a gout flare, but this incidence was similar between groups. The gout flares became less common toward the end of the study (8–11% incidence at weeks 49–52) supporting the usual UK practice of providing prophylaxis for 3–6 months when prescribing urate-lowering therapy [Jordan et al. 2007].

The incidence of adverse events was similar in all three treatment groups and included liver function abnormality, diarrhoea, headache, and joint and muscle pain. The four deaths which occurred were judged unrelated to the study drugs and the numbers were not statistically significant between groups.

The obvious critiques of the FACT study were the failure to include those with a serum creatinine > 133 μmol/l, as this patient group would comprise a significant proportion of a usual gout patient cohort and that the allopurinol dose was not uptitrated as per usual practice. Uptitrating the allopurinol would reasonably have been expected to result in better reduction in serum urate [Perez-Ruiz et al. 1998].

The Allopurinol and Placebo-Controlled Efficacy Study of Febuxostat (APEX) [Schumacher et al. 2008] was a phase III, randomized, double-blind, randomized, 28-week study to compare the safety and efficacy of febuxostat with placebo and allopurinol for the treatment of gout with a baseline serum urate >0.48 mmol/l (>8.0 mg/dl). APEX aimed to seek confirmation of the findings of FACT and to include patients with impaired renal function that APEX had excluded (n = 40). Overall, subjects were randomized in a 2:2:1:2:1 ratio to febuxostat 80 mg (n = 267), febuxostat 120 mg (n = 269), febuxostat 240 mg (n = 134), allopurinol (n = 268) dose dependent on renal function, or placebo (n = 134). Patients were given NSAID or colchicine prophylaxis for the first 8 weeks of double blind treatment.

The primary endpoint in APEX, as in FACT, was a serum urate of <0.36 mmol/l (6.0 mg/dl) at the last 3-monthly measurements. The primary endpoint was reached in 48% of the febuxostat 80 mg group, 65% of the febuxostat 120 mg group, 69% of the febuxostat 240 mg group, 22% of the allopurinol group and 0% of the placebo group. Febuxostat was significantly more likely, at any dose, to result in the achievement of the primary endpoint than allopurinol or placebo. Again, a statistically higher incidence of gout flare was seen in the first 8 weeks with those treated with febuxostat 120 mg (36%) but also with 240 mg (46%) than those receiving febuxostat 80 mg (28%), allopurinol (23%) or placebo (20%). The incidence of flare for the rest of the trial then became similar between groups.

No significant differences in the number of tophi were observed between groups, with the exception of a mean percentage decrease in the number of tophi seen with febuxostat 120 mg versus placebo at the end of the study.

Adverse events were generally found to occur with similar frequency across the treatment arms and were mild-to-moderate in severity with upper respiratory tract infection, musculoskeletal and connective tissue signs and symptoms, diarrhoea, joint signs and symptoms, headaches and liver function abnormalities the most commonly seen. Diarrhoea and dizziness were more commonly seen in the febuxostat 240 mg group than in the lower-dose groups. There were no deaths in the study.

The data in APEX corroborated the novel findings of FACT. Again, the lack of the use of a titration regime of allopurinol may have led to the poorer-than-expected performance of allopurinol as urate-lowering therapy. APEX did have the benefit of the inclusion of patients with renal impairment and though this group was small (40), those treated with febuxostat were statistically more likely to achieve the primary endpoint than those receiving placebo or allopurinol.

The CONFIRMS [Becker et al. 2010] study was a phase III, double blind, randomized controlled trial to further examine the comparative efficacy and safety of febuxostat compared with that of allopurinol in subjects with mild-to-moderate renal impairment for the treatment of gout with a baseline serum urate >0.48 mmol/l (8.0 mg/dl). At least 35% of subjects were to have mild-or-moderate renal impairment with thresholds of estimated creatinine clearance of 60–89 ml/min or 30–59 ml/min, respectively.

Subjects were randomized to receive febuxostat 40 mg (n = 757), febuxostat 80 mg (n = 756) or allopurinol 200/300 mg (n = 755). Subjects received prophylaxis for the duration of 6 months of the study, of colchicine 600 mg per day or naproxen 250 mg twice a day.

The primary endpoint was the proportion of subjects with a serum urate of <360 μmol/l at the final visit. The authors stated that a major aim was to demonstrate noninferiority of febuxostat 40 mg against allopurinol. This endpoint was reached in 45.2% of the febuxostat 40 mg group, 67.1% in the febuxostat 80 mg group and 42.1% in the allopurinol group. Those in the febuxostat 80 mg group were statistically more likely to reach the endpoint than in the other groups and febuxostat 40 mg was noninferior to allopurinol. In the subjects with renal impairment, febuxostat 80 mg was significantly more effective than febuxostat 40 mg or allopurinol.

Rates of gouty flare requiring treatment occurred in 10–15% of subjects in all treatment groups over the first 2 months, but then declined over the subsequent 4 months.

The occurrence of adverse events were similar amongst treatment arms and the majority were mild-to-moderate in severity. The adverse event rates were similar in those with renal impairment as those in the rest of the study.

The CONFIRMS study served to give further evidence regarding the efficacy and safety of febuxostat, particularly in those patients with renal impairment. It showed equivalence between febuxostat 40 mg and the allopurinol group overall but also, that in those with renal impairment, febuxostat 40 mg was more effective than allopurinol; 80 mg of febuxostat was superior to the other groups. As in FACT and APEX, no titration of allopurinol was included.

A further randomized controlled trial, lasting 28 weeks, published in 2014, aimed to compare the urate-lowering efficacy and safety of febuxostat and allopurinol in Chinese patients with hyperuricaemia and gout [Huang et al. 2014]. In the trial, 516 subjects with a serum urate > 0.48 mmol/l (8.0 mg/dl) were randomized to receive either febuxostat 40 mg, febuxostat 80 mg or allopurinol 300 mg (with 172 subjects in each group). Those taking a previous urate-lowering therapy had a 2-week washout period prior to commencing the study. Those with a serum creatinine > 135 μmol/l were excluded. Colchicine 500 mg per day or meloxicam 7.5 mg per day was given as gout flare prophylaxis for the first 8 weeks of the trial.

As in APEX and FACT, the primary endpoint was a serum urate of <0.36 mmol/l (6.0 mg/dl) at the last 3-monthly measurements. The primary endpoint was reached in 27.33% of the febuxostat 40 mg, 44.77% of the febuxostat 80 mg and 23.84% of the allopurinol group. Those in the febuxostat 80 mg group were statistically more likely to reach the primary endpoint than in the febuxostat 40 mg or allopurinol 300 mg group. The urate-lowering efficacy of the febuxostat 40 mg was noninferior to the allopurinol 300 mg group. The secondary endpoint of tophi reduction was not met, as no significant change in the number of tophi was seen between the three groups. There was no difference between the groups in the rates of those requiring treatment for acute gout flares from weeks 9–28, but the authors have not reported the rates of flare in weeks 0–8, which one would expect from previous experience to have been higher in the febuxostat 80 mg group in this time period due to the rapid urate-lowering effect.

The incidence of adverse events was similar between groups and the majority were mild−moderate in severity. There were no deaths in the study. The most frequently seen adverse events in the febuxostat groups were abnormal liver function, hyperlipidaemia, raised creatinine and gastrointestinal disturbance. The authors classified gouty flare as ‘expected gout manifestation’ rather than as adverse events, thus explaining the paucity of joint or soft tissue adverse events so commonly reported in earlier trials. None of the serious events were judged related to the study drug, except for the skin rash caused by allopurinol.

A Japanese, phase III, placebo-controlled double-blinded study published in 2011 aimed to evaluate the efficacy and safety of febuxostat 20 mg or 40 mg in patients with hyperuricaemia including some with gout over an 8-week trial period [Yang et al. 2010]. Patients did not receive prophylaxis against gouty flare and those in the febuxostat groups were given an ‘introductory dose’ of 10 mg per day for 2 weeks. Only a proportion of the patients had a history of clinical gout and it is not mentioned what diagnostic criteria were used to define this. None of the 19 patients with gout who received placebo, 52.9% of the 17 receiving 20 mg febuxostat and 92.9% of those receiving 40 mg febuxostat reached a serum urate of <0.36 mmol/l (6 mg/dl) which was the primary variable. Any subgroup analysis of adverse events in the gout group is not presented. This trial serves to show efficacy of febuxostat in reduction of serum urate in the short term in patients with hyperuricaemia, including some with gout. However, the doses of febuxostat used are not usual in the UK. The late phase II dose-response study by the same group [Kamatani et al. 2011] had found the percentage of patients attaining target serum urate differed little between the groups receiving 40–80 mg of febuxostat per day, which presumably informed the decision to use the doses of 20 mg and 40 mg in the phase III work in this group of patients. In addition, there was no comparison with the gold standard urate-lowering therapy (allopurinol) and there was no adverse-event analysis in the subgroup with gout. Thus, this is less useful to inform UK practice than the articles discussed above.

Extension studies

EXCEL [Becker et al. 2009] was a long-term open-label extension study of 1086 patients for up to 40 months for patients completing FACT and APEX. It aimed to determine the long-term urate-lowering efficacy and safety in those receiving febuxostat or allopurinol in subjects with hyperuricaemia and gout. Patients were assigned in a 2:2:1 ratio to receive febuxostat 80mg, febuxostat 120 mg, or allopurinol 100 mg or 300 mg. Additionally, 8 weeks of colchicine 600 mg per day or naproxen prophylaxis was given. Patients could be reassigned during months 1–6 to receive another urate-lowering therapy if target serum urate was not reached. Over 80% of the febuxostat-treated patients reached target after 1 month and this was sustained above 80% for the rest of the study. Flare was seen commonly after week 8 when prophylaxis was withdrawn but rates of flare diminished over time and were reported in <4% of subjects after 18 months of urate-lowering therapy. Resolution of the index tophus was achieved by 46%, 36%, and 29% of subjects maintained on febuxostat 80 mg, febuxostat 120 mg, and allopurinol, respectively. The total adverse events and serious adverse events including cardiovascular events (which were all seen in those with existing cardiovascular disease and felt unrelated to the drugs) were similar across all groups adjusted for patient−year exposure to drug.

FOCUS [Schumacher et al. 2009] followed the first phase II trial discussed above [Becker et al. 2005b]. It was a 5-year extension study to show efficacy at lowering and maintaining serum urate < 360 mmol/l (6.0 mg/dl). In the trial, 116 patients received febuxostat 80 mg per day for the first 4 weeks of the study, whilst also receiving 600 mg twice a day of colchicine as prophylaxis. The febuxostat dose could then be titrated up to three times, up to week 24, to maintain serum urate between 0.18–0.36 mmol/l (3.0–6.0 mg/dl) but subjects needed to be on a stable dose for 4 weeks by week 28. At 5 years, 93% of the 58 subjects who had completed the study had a serum urate < 0.36 mmol/l (6.0 mg/dl). Rates of attaining the primary endpoint were similar, whether or not renal impairment was present. Self-reported gout flare was seen commonly early in the study; these flares tended to occur at increased frequency in subjects with baseline tophi, likely reflecting the long time taken to fully mobilize the large burden of uric acid crystals in these patients. None of the subjects reported a gout flare by year 5 of the study compared with the peak of 23% of subjects by month 2. Of the 26 with tophi at baseline, 69% had resolved by the end of the study. Febuxostat was considered safe with the majority of adverse events mild to moderate. None of the serious adverse events were judged by the authors to be related to febuxostat, but this was a subjective finding and there was no placebo or allopurinol control to compare adverse event rates with. There was also a high discontinuation rate of 50% in the study.

Interactions

Although the interaction of febuxostat with azathioprine or mercaptopurine has not been studied, the concurrent use of these drugs is not recommended. This is because xanthine oxidase inhibition shifts the metabolism of azathioprine and mercaptopurine toward the production of metabolites which can cause rapid and severe leucopaenia [Kennedy et al. 1996; Venkat Raman et al. 1990].

An interaction study in healthy subjects showed there to be no effect on the pharmacokinetics or safety of 400 mg theophylline when febuxostat 80 mg per day was given concomitantly [European Medicines Agency, 2012].

Safety and tolerability

Studies to date suggest febuxostat to be safe and well tolerated. The overall rates of adverse events reported in the placebo arm of APEX, the only phase III work to contain a placebo arm, were similar to those in the febuxostat groups. As with all drugs, acute hypersensitivity reactions have been reported [European Medicines Agency, 2012] and as with the administration of all drugs, this should be borne in mind when assessing a patient recently commenced on any new medication. Hypersensitivity to allopurinol is not a contraindication to febuxostat. Although it is recommended that patients to be treated with febuxostat have liver function checked prior to therapy and rechecked thereafter on a ‘discretionary’ basis, while liver function derangement was seen more commonly in the febuxostat groups in APEX, none of the trials discussed showed combined transaminase and bilirubin increases, aside from those attributable to biliary tract disease.

There has been concern regarding the potential cardiovascular risks associated with febuxostat based on data from phase III trials and postmarketing experience. In pooled data from FACT and APEX, a numerically higher incidence of cardiovascular events (Anti-Platelet Trialists’ Collaboration composite endpoint of nonfatal MI, nonfatal stroke or cardiovascular death) was seen in patients receiving febuxostat, rather than allopurinol. This was not statistically significant and was not replicated in the CONFIRMS study. Nonetheless, in the UK, febuxostat is not recommended for those with ischaemic heart disease or congestive cardiac failure. However, it is as yet unclear whether the cardiovascular events seen in this work are related to the high prevalence of traditional risk factors such as male gender, increasing age, hypertension and diabetes in patients with hyperuricaemia and gout. To further investigate this association, two large-scale trials (FAST and CARES) to investigate this are underway [White et al. 2012; MacDonald et al. 2014].

Febuxostat in the management of gout in the UK

Febuxostat is undoubtedly effective at reducing serum urate, both in the short and long term. When used for a long period, it is also effective at reducing the burden of tophi and in abolishing gouty flare. The lack of need for dose adjustment in those with mild−moderate renal impairment makes it an attractive choice in that group of patients.

However, it is currently contraindicated in those with ischaemic heart disease and congestive cardiac failure, which, of course, occur commonly in the gout population.

There is also a question of the relevance of the trials discussed to optimal management of gout in the UK. None of the trials comparing febuxostat with allopurinol used a gold standard allopurinol regime, whereby patients are uptitrated to achieve target serum urate, although this strategy is suggested by multiple guidelines of international significance. We appreciate that one potential justification for these study designs is that failure to adopt this strategy of allopurinol dose titration to target is somewhat representative of real-world practice where patients commonly remain on subtherapeutic allopurinol doses and no reassessment of urate is made, or where it is, no uptitration of therapy occurs [Roddy et al. 2007]. Where 300 mg of allopurinol was the maximum dose in these studies, it is widely acknowledged that at these doses, <50% of patients will reach target with some patients needing up to 900 mg per day [Jordan et al. 2007]. A double-blinded trial comparing an uptitrating allopurinol regime with febuxostat would be desirable (if unlikely to be undertaken) to demonstrate whether either regime would be superior to the other.

In addition, the BSR [Jordan et al. 2007] suggests a serum urate target of <0.3 mmol/l (5 mg/dl) whereas the primary endpoints in the studies were <0.36 mmol/l (6 mg/dl) making it difficult to generalize conclusions to the UK population.

Conclusion

We conclude that febuxostat is a highly effective serum urate-lowering drug. It should be given to achieve serum urate <0.3 mmol/l (5.0 mg/dl). In the long term, this will result in a reduction of gout flares and in tophi resolution. It should also prevent the development of chronic gouty arthropathy if used early enough in the disease course.

Febuxostat is a potent serum urate-lowering agent and therefore its use is associated with significant rates of gouty flare on initiation of therapy. Patients should be counselled with regard this. Therefore, we recommend that the use of febuxostat should be accompanied by an adequate period of colchicine or NSAID prophylaxis (the choice of which should be informed by the comorbidities of the patient). The BSR recommends colchicine 500 mg twice a day for up to 6 months or, if this is not tolerated, an NSAID or COXIB for a more limited period of 6 weeks [Jordan et al. 2007]. We would suggest the coprescription of appropriate proton pump inhibitor cover with NSAID or COXIB.

As febuxostat has not been proven superior to a gold-standard escalating regime of allopurinol, we recommend its use only for those intolerant of allopurinol, or in those in whom renal function precludes optimal dose escalation to achieve target serum urate. We cannot as yet advocate an alternate dosing regimen of allopurinol for those with significant renal impairment until there are further efficacy and safety data available but would advocate in patients with chronic kidney disease that a gradual dose escalation of allopurinol is employed with close renal-blood monitoring dictating uptitration of allopurinol dosing.

Patients with ischaemic heart disease and congestive cardiac failure (who often have coexistent renal impairment) account for a large proportion of our patients with gout that are challenging to control. We therefore wait with interest, the findings of FAST and CARES [White et al. 2012; MacDonald et al. 2014] to investigate the cardiovascular safety of febuxostat. Should these be favourable, we would anticipate a marked increase in the use of febuxostat in UK practice.