Febuxostat for treating allopurinol-resistant hyperuricemia in patients with chronic kidney disease

Abstract

Background: Availability of the novel xanthine oxidase inhibitor febuxostat, which has multiple excretion pathways, enables investigation of the significance of serum uric acid control on renal function in patients with chronic kidney disease (CKD). Methods: This was an exploratory, retrospective, observational study conducted at a single Japanese center. Serum uric acid concentrations and serum creatinine levels in the 6 months before and after the start of febuxostat treatment were collected for CKD patients switched from allopurinol after failing to achieve serum uric acid concentrations ≤6.0 mg/dL. Results: Evaluable data were available for 60 patients, 67% of whom had advanced CKD (eGFR <30 mL/min/1.73 m²). Mean dose of febuxostat was 15.9 (± 8) mg/day. Mean serum uric acid concentration decreased from 8.4 (±1.4) mg/dL at baseline to 6.2 (±1.2) mg/dL at 6 months; 47.5% of patients achieved a level ≤6.0 mg/dL. The change from baseline in eGFR was positive at all time points during febuxostat treatment and the increase of 2.3 (±5.6) mL/min/1.73 m² at 6 months was significant (p = 0.0027). Whereas the eGFR slope was negative during allopurinol treatment, it became positive after the switch to febuxostat. The change in eGFR slope before and after febuxostat treatment was significant for all patients (p < 0.01), for male patients (p < 0.05), and for patients with a baseline eGFR of <15 mL/min/1.73 m² (p < 0.05). Conclusions: In patients with CKD, febuxostat reduces serum uric acid concentrations effectively and may suppress the progressive decline in renal function.

• Allopurinol
• chronic kidney disease
• febuxostat
• hyperuricemia

Introduction

An estimated 13.3 million Japanese people – or 1 in 8 adults – have chronic kidney disease (CKD).1 Treatment of CKD is aimed at slowing the progression to end-stage kidney disease (ESKD) and the onset of cardiovascular disease (CVD). To this end, multimodal treatment ranging from lifestyle and dietary guidance to blood pressure, glucose, and lipid control through to correction of renal anemia and abnormalities in bone and mineral metabolism is essential to prevent or delay a series of pathologies.1 Recently, serum uric acid has become a target of control as per the 2012 revision of the Japanese Society of Nephrology CKD treatment guideline.2

Hyperuricemia is defined as a serum uric acid concentration greater than 7.0 mg/dL.3 As the dissolution limit of urate in body fluids is around 6.4 mg/dL, persistent hyperuricemia causes formation and deposition of urate crystals in body tissues. The typical resultant disease is gouty arthritis; however, urolithiasis and gouty nephropathy are also known to occur. In gouty nephropathy, crystals deposit in the renal medulla, inducing chronic interstitial nephritis. Moreover, it has recently been considered that oxidative stress due to hyperuricemia induces arteriosclerosis in association with hypertension and glucose/lipid metabolism disorder, resulting in renal hypertension and reduced renal function as pathologies unrelated to urate deposition. Nearly a decade ago already, Iseki and colleagues reported an association between an increased risk of ESKD and serum uric acid concentrations of ≥7 mg/dL in males and ≥6 mg/dL in females.4

Few studies to date have investigated the significance of treating hyperuricemia on renal function, likely due to the well-known difficulties of using conventional urate-lowering drugs in patients with reduced renal function. Allopurinol suppresses uric acid production by inhibiting xanthine oxidase, but it has a purine structure and is known to affect other nucleic acid metabolic enzymes. Metabolism of allopurinol by xanthine oxidase produces the active metabolite, oxypurinol, which is excreted almost entirely by the kidneys. Oxypurinol plasma concentrations are easily elevated to the toxic range in patients with impaired renal function; this often precludes administering allopurinol at a dose necessary to adequately reduce serum uric acid concentrations. In addition, uricosuric agents such as benzbromarone are not recommended for patients with reduced renal function due to an attenuated effect.

Febuxostat is a novel xanthine oxidase inhibitor introduced in Japan in May 2011. It does not contain purine in its chemical structure and is a highly selective, potent, and continuous inhibitor of xanthine oxidase via a new mechanism of inhibition.5 Multiple excretion pathways permit its use even in patients with reduced renal function. Febuxostat has shown good efficacy with regard to patients achieving a serum uric acid target level of ≤6.0 mg/dL.6,7 In this regard a number of well-controlled clinical trials such as APEX,8 FACT,9 and EXCEL10 have shown that febuxostat (80–240 mg/day) was more effective than allopurinol (100–300 mg/day) in lowering and maintaining serum urate levels below 6 mg/dL in patients with hyperuricemia and gout. Moreover, its availability enables investigation of the potential impact of active intervention for hyperuricemia on renal function.

The purpose of this exploratory study was to evaluate the urate-lowering activity of febuxostat and its effect on renal function in patients with CKD.

Materials and methods

This was a retrospective observational exploratory study performed in patients who had attended the Nippon Medical School Musashikosugi Hospital between July 2011 and November 2012. The study population involved patients with CKD (estimated glomerular filtration rate [eGFR] <60 mL/min/1.73 m²) who had failed treatment with allopurinol (serum uric acid concentrations remained ≥6.0 mg/dL) and who had been treated with febuxostat for 6 months or longer. The trial was performed in accordance with Japanese Good Clinical Practice and Good Postmarketing Surveillance Practice guidelines, and in line with the ethical principles set out in the Declaration of Helsinki.

Data regarding serum uric acid concentrations and serum creatinine levels in the 6-month periods before and after the start of febuxostat treatment were extracted from patients’ electronic medical records. The usual laboratory methods employed by the hospital were used throughout the study. Estimated GFR was calculated using a formula based on serum creatinine developed by the Japanese Society of Nephrology for the Japanese population [eGFR (mL/min/1.73 m²) = 194 × Cr^−1.094× Age^−0.287 (×0.739 if women).1 Based on monthly eGFR values a regression line was calculated for each subject to determine the rate of change in eGFR per 6-month period, and this was defined as the slope of the linear relationship between renal function and time for each individual.11

Statistical analysis

Changes in serum uric acid concentrations and eGFR in the 6-month periods before and after the switch to febuxostat were the primary assessment parameters. Each measurement was presented as a mean ± standard deviation and a paired t-test was used to determine statistical significance. The relationship between change in serum uric acid concentrations and change in eGFR at 6 months after the switch to febuxostat was investigated using Pearson’s correlation coefficient and Spearman’s rank correlation coefficient. Estimated GFR slopes in the 6 months before and after the switch to febuxostat were compared and the paired t-test was used to determine statistical significance. For all statistical analyses p < 0.05 was considered statistically significant. Analysis software was SAS version 9 in a Windows 7 operating system (SAS® Software, Cary, NC).

Results

Patient characteristics

Sixty patients were able to be followed for 6 months after switching from allopurinol to febuxostat. Baseline characteristics are presented in Table 1. Most patients were male (73%) and the mean age of the cohort was 63.1 (±16.6) years. Mean serum uric acid concentration before switching to febuxostat was 8.4 (±1.4) mg/dL (range: 6.2–13.1 mg/dL). Two-thirds of patients (n = 40) had advanced CKD (eGFR <30 mL/min/1.73 m²). Primary conditions underlying renal disease were hypertension (n = 15), chronic glomerulonephritis (n = 21), and diabetic nephropathy (n = 10), while other less common causes were responsible for the remaining 14 cases (Table 1). Four patients had a history of gouty kidney disease of mean duration 6.5 (±2.1) years. In terms of concomitant medications, 45 patients were being treated with an angiotensin converting enzyme (ACE) antagonist or angiotensin II receptor blocker (ARB), none were prescribed a non-steroidal anti-inflammatory drug (NSAID) on a regular basis, and alcohol was used only occasionally (no heavy drinkers were included in this cohort).

Table 1. Baseline characteristics of the study population.

Baseline characteristics	All patients n = 60	Male patients n = 44 (73%)	Female patients n = 16 (27%)
Age (years)
Mean (±SD)	63.1 ± 16.6	62.8 ± 16.4	63.9 ± 17.7
Range	29–98	30–89	29–98
Serum uric acid concentration (mg/dL)
Mean (±SD)	8.4 ± 1.4	8.5 ± 1.4	8.2 ± 1.6
Range	6.2–13.1	6.3–13.1	6.2–12.1
<7 mg/dL (n, %)	6 (10.0)	4 (9.1)	2 (12.5)
7–8 mg/dL (n, %)	16 (26.7)	10 (22.7)	6 (37.5)
8–9 mg/dL (n, %)	25 (41.7)	20 (45.5)	5 (31.3)
≥9 mg/dL (n, %)	13 (21.7)	10 (22.7)	3 (18.8)
eGFR (ml/min/1.73 m^2)
Mean ( ± SD)	27.2 ± 19.2	29.8 ± 20.7	20.2 ± 12.0
Range	3.1–89.8	4.0–89.8	3.1–43.7
<15 mL/min/1.73 m^2 (n, %)	18 (30.0)	12 (27.3)	6 (37.5)
15–29 mL/min/1.73 m^2 (n, %)	22 (36.7)	15 (34.1)	7 (43.8)
30–44 mL/min/1.73 m^2 (n, %)	11 (18.3)	8 (18.2)	3 (18.8)
45–59 mL/min/1.73 m^2 (n, %)	4 (6.7)	4 (9.1)	0
≥60 mL/min/1.73 m^2 (n, %)	5 (8.3)	5 (11.4)	0
Underlying morbidity
Hypertension	15 (25.0)	11 (25.0)	4 (25.0)
Chronic glomerulonephritis (n, %)	21 (35.0)	16 (36.4)	5 (31.3)
IgA nephropathy (n, %)	12 (20.0)	8 (18.2)	4 (25.0)
Membranous nephropathy (n, %)	7 (11.7)	6 (13.6)	1 (6.3)
Focal segmental glomerulosclerosis (n, %)	1 (1.7)	1 (2.3)	0
Biopsy not performed (n, %)	1 (1.7)	1 (2.3)	0
Diabetic nephropathy (n, %)	10 (16.7)	6 (13.6)	4 (25.0)
Gouty nephropathy (n, %)	4 (6.7)	4 (9.1)	0
Polycystic kidney (n, %)	4 (6.7)	3 (6.8)	1 (6.3)
Interstitial nephritis (n, %)	2 (3.3)	1 (2.3)	1 (6.3)
Rheumatoid arthritis (n, %)	2 (3.3)	1 (2.3)	1 (6.3)
MRSA nephropathy (n, %)	1 (1.7)	1 (2.3)	0
Renal cell carcinoma (n, %)	1 (1.7)	1 (2.3)	0

Abbreviations: MRSA, methicillin-resistant Staphylococcus aureus; SD, standard deviation.

Febuxostat was administered at a daily dose of 10 mg in 33 patients, 20 mg in 21 patients, 30 mg in 3 patients, and 40 mg in 3 patients. The mean dose in all patients was 15.9 (±8) mg/day. The mean allopurinol dose prior to the switch was 71.3 (±29.5) mg/day.

Serum uric acid concentrations

Changes in serum uric acid concentrations in the 6-month periods before and after the switch from allopurinol to febuxostat are shown in Figure 1. The mean serum uric acid concentration was significantly (p < 0.001) reduced from baseline at 1 month after the switch to febuxostat and remained significantly lower throughout treatment. At 6 months, the mean serum uric acid concentration was 6.2 (±1.2) mg/dL. Nearly half the population (47.5%) achieved a serum uric acid concentration ≤6.0 mg/dL. Reduction in the mean serum uric acid concentration was satisfactory irrespective of gender (Figure 2a) or baseline renal function (Figure 2b).

Figure 1. Change in serum uric acid concentration (mean ± SD) before and after the start of febuxostat treatment (number in parentheses indicates number of patients). ***p < 0.001 versus Month 0 (paired t-test).

Figure 2. Change in serum uric acid concentrations (mean ± SD) during febuxostat treatment stratified by (a) gender and (b) baseline eGFR (mL/min/1.73 m²). ***p < 0.001 versus Month 0 (paired t-test).

eGFR

After switching to febuxostat, there was a positive change from baseline in the mean eGFR at all time points (Table 2). The increase of 2.3 (±5.6) mL/min/1.73 m² at 6 months was significant (p = 0.0027) versus baseline. No clear correlation between the change in serum uric acid concentrations and change in eGFR after 6 months of febuxostat treatment was found overall or in most patient subgroups stratified by gender or baseline eGFR (Table 3). A significant inverse correlation (r = −0.70; p = 0.0013) was detected only in patients whose baseline eGFR was <15 mL/min/1.73 m² (Table 3). Subsequent examination of eGFR before and after the switch to febuxostat in this particular patient group showed that, in contrast to the steady decline in renal function during allopurinol treatment, there was a tendency towards improvement in eGFR during febuxostat treatment (Figure 3).

Figure 3. Change in eGFR (mean ± SD) before and after the start of febuxostat treatment in patients with a baseline eGFR <15 mL/min/1.73 m² (number in parentheses indicates number of patients). *p < 0.05 versus Month 0 (paired t-test).

Table 2. Change from baseline in eGFR (mL/min/1.73 m²) after the start of febuxostat treatment.

	Total			Male			Female
	n	Mean ± SD	p Value*	n	Mean ± SD	p Value*	n	Mean ± SD	p Value*
Baseline (Month 0)	60	27.2 ± 19.2	–	44	29.8 ± 20.7	–	16	20.2 ± 12.0	–
Month 1	54	1.2 ± 4.1	0.0416	40	0.9 ± 4.1	0.1672	14	1.8 ± 4.0	0.1076
Month 2	59	1.8 ± 6.0	0.0252	44	1.9 ± 6.7	0.0650	15	1.5 ± 3.6	0.1294
Month 3	53	0.7 ± 4.3	0.2308	38	0.6 ± 4.9	0.4638	15	1.1 ± 2.4	0.1082
Month 4	59	0.6 ± 5.7	0.4354	44	0.6 ± 6.2	0.5420	15	0.6 ± 3.7	0.5525
Month 5	52	0.8 ± 4.4	0.2008	38	0.7 ± 4.7	0.3588	14	1.0 ± 3.7	0.3165
Month 6	58	2.3 ± 5.6	0.0027	42	2.5 ± 5.8	0.0074	16	1.7 ± 5.1	0.1998

Symbols: *versus Month 0 (paired t-test).

Table 3. Correlation between the change in serum uric acid concentrations and change in eGFR at 6 months after the start of febuxostat treatment.

Patient group	Pearson correlation coefficient (r)	p Value
Total (n = 58)	−0.10	0.4747
Male (42)	−0.02	0.8860
Female (16)	−0.29	0.2696
eGFR at Month 0: <15 mL/min/1.73 m^2 (18)	−0.70	0.0013
eGFR at Month 0: 15–30 mL/min/1.73 m^2 (21)	0.21	0.3694
eGFR at Month 0: 30–45 mL/min/1.73 m^2 (11)	−0.52	0.1022
eGFR at Month 0: 45–60 mL/min/1.73 m^2 (3)	−0.97	0.1685
eGFR at Month 0: ≥60 mL/min/1.73 m^2 (5)	0.71	0.1835

Accordingly, eGFR slopes in the 6-month periods before and after the switch to febuxostat were evaluated to determine the effect of active reduction of serum uric acid concentrations on renal function. Whereas the eGFR slope was negative during allopurinol treatment, it became positive after the switch to febuxostat (Figure 4) and was positive irrespective of gender (Figure 5a) or baseline renal function (Figure 5b). The change in eGFR slope before and after febuxostat treatment was significant for all patients (p < 0.01), for male patients (p < 0.05), and for patients with a baseline eGFR <15 mL/min/1.73 m² (p < 0.05).

Figure 4. Change in eGFR slope (mean ± SD) before (a) and after (b) the start of febuxostat treatment (number in parentheses indicates number of patients). *p < 0.05 versus Month 0 (paired t-test).

Figure 5. Change in eGFR slope (mean ± SD) before and after febuxostat treatment stratified by (a) gender and (b) baseline eGFR (mL/min/1.73 m²). *p < 0.05 and **p < 0.01 versus before treatment (paired t-test).

Adverse events

A slight increase in transaminase levels was recorded in a single patient after switching to febuxostat, but the event was not considered clinically significant and febuxostat treatment was continued as before.

Discussion

Hyperuricemia is frequently associated with hypertension, diabetes, dyslipidemia, and other metabolic disorders.1 As uric acid is cleared through the kidneys, a decline in renal function can lead to hyperuricemia especially in elderly persons long affected by these lifestyle diseases. In turn, it is becoming increasingly accepted that hyperuricemia contributes to the development and progression of renal pathology by both direct and indirect mechanisms. In the current study, retrospective analyses were performed to investigate the effects of switching from the conventional urate-lowering agent, allopurinol, to the novel urate-lowering agent, febuxostat, on uric acid control and renal function in patients with CKD.

Febuxostat produced significant reductions in mean serum uric acid concentrations from Month 1 onwards. Nearly half the patient sample (47.5%) reached the target serum uric acid concentration of ≤6.0 mg/dL with a 10–20 mg dose; even greater efficacy could be expected with higher doses. Importantly, the slow but steady decline in eGFR observed with allopurinol in the 6 months prior to the switch was halted by febuxostat and there was a tendency towards improvement. Although data were not presented due to small sample sizes, no major differences were observed in the efficacy of febuxostat according to presence or type of underlying disease (e.g., hypertension, glomerulonephritis, etc). Taken together, these findings suggest that febuxostat may be useful in the maintenance of renal function in patients with CKD.

A study from the US which examined the long-term (5-year) effects of febuxostat on eGFR found an inverse correlation between the quantitative reduction from baseline in serum uric acid concentrations and maintenance or improvement of eGFR.12 In the current study, this relationship was detected only in the subgroup with a baseline eGFR <15 mL/min/1.73 m². Studies over a longer term and with larger sample sizes should help confirm the inverse association between the magnitude of reduction in serum urate levels and improvement in renal function.

Uric acid is known to be a potent antioxidant. However, studies in vivo and in vitro have suggested that uric acid may induce oxidative stress and inflammation in vascular endothelial cells leading to reduced production of nitric oxide13 or that it may induce renal arteriopathy or tubulointerstitial fibrosis through activation of the renin–angiotensin system.14 Both sets of results suggest that elevation of serum uric acid concentrations can lead to renal impairment, and support the establishment of uric acid as a therapeutic target to prevent the onset and progression of CKD.2 Indeed, support for uric acid as a target of therapy in CKD derives from several sources. In an investigation of 167 patients who underwent renal biopsy, higher serum uric acid concentrations were associated with a greater degree of hyaline degeneration and arterial wall thickening in the renal artery.15 Elsewhere, high-dose allopurinol was shown to exhibit hypotensive activity in hyperuricemic adolescents with newly-diagnosed hypertension.16 Hiramitsu and colleagues reported blood pressure improvement in patients with hyperuricemia treated for 12–16 weeks with febuxostat under general practice conditions.17 Reducing serum uric acid concentrations may improve vascular endothelial dysfunction, leading to functional and structural improvement of blood vessels. Although the effect of febuxostat on blood pressure was not analyzed in the current study, this same mechanism may underlie the improvement in eGFR observed with febuxostat.

Guidelines for treatment of hyperuricemia/gout recommend a target serum uric acid concentration of ≤6.0 mg/dL,3 or even <5.0 mg/dL,18 to achieve durable control of the signs and symptoms of gout. These recommendations stem from the need to maintain serum uric acid concentrations below the dissolution limit of extracellular fluid. In the current study, it was considered that removing urate deposits in the kidney was one of the factors underlying the tendency towards improvement in eGFR seen with febuxostat.

At our center we have experienced other cases in which urate crystals identified on kidney sonogram following insufficient reduction of serum uric acid concentrations with allopurinol disappeared when levels were maintained below 6.0 mg/dL by treatment with febuxostat. In the study of Goicoechea and colleagues, 113 CKD patients with eGFR <60 mL/min/1.73 m² were randomly assigned to receive allopurinol 100 mg/day or continue with usual therapy. After a mean follow-up of 24 months, serum uric acid concentrations were reduced and the decline in eGFR was suppressed in the group receiving allopurinol compared to the control group. Serum uric acid concentrations were reduced from 7.8 to 6.0 mg/dL in the allopurinol group but remained unchanged (7.3 to 7.5 mg/dL) in the control group.19 These findings highlight the benefits that can be achieved by reducing and maintaining serum uric acid concentrations at around 6.0 mg/dL in patients with impaired renal function.

It has been demonstrated clinically that febuxostat inhibits xanthine oxidase more strongly than allopurinol and has a potent uric acid-reducing effect. In addition, its inhibitory effect on xanthine oxidase-dependent reactive oxygen species formation was shown to be 1000-fold greater than that of allopurinol.20 Recently, Isaka and co-workers reported that febuxostat showed inhibitory activity on oxidative stress markers including nitrotyrosine and thiobarbituric acid reactive substances (TBARS), producing an anti-inflammatory and renal protective effect in animal models with renal ischemia-reperfusion injury and ureteral ligation.21,22 Thus, the renal protective effect of febuxostat may not only be due to its potent uric acid-reducing effect, but also to inhibition of oxidative stress through potent inhibition of xanthine oxidase in pathological conditions related to reduced renal function.

As this was an exploratory, retrospective study it has a number of limitations. In particular, the number of patients with data available for analysis was small for some subgroup analyses. As such, the findings should be interpreted with caution. The observation period of 6 months was relatively short and precludes extrapolating the results over the longer term. Nevertheless, it was highly encouraging to demonstrate the uric acid-lowering efficacy of febuxostat and the trend towards slowing the progression of renal dysfunction in patients with CKD. Prospective, long-term, large-scale, randomized comparative studies are required to further elucidate the significance of actively treating hyperuricemia in patients with CKD.

Declaration of interest

The authors report no conflict of interest. The authors alone are responsible for the content and writing of the paper.

Acknowledgments

Assistance with English language was kindly provided by Content Ed Net.