The effects of topiroxostat on vascular function in patients with hyperuricemia

Abstract

Xanthine oxidoreductase (XOR) inhibitors, such as allopurinol and febuxostat, inhibit the catalysis of serum uric acid (SUA) synthesis. In doing so, they are thought to improve vascular endothelial function in patients with hyperuricemia and cardiovascular risk by reducing increases in SUA and reactive oxygen species levels. We performed a retrospective cohort study to evaluate the effects of topiroxostat, a novel XOR inhibitor, on vascular function measured by flow‐mediated dilation (FMD) on ultrasonography. In total, 23 patients with hyperuricemia were enrolled. After approximately 8 weeks, topiroxostat was associated with a significant increase in the peak percentage change in diameter (∆FMD) from 4.53% ± 2.09% to 5.54% ± 3.08% (P = .045). It also significantly reduced the SUA levels from 7.31 ± 1.43 to 5.44 ± 1.11 mg/dL (P < .001). Although further studies are needed to validate these results, it appears that topiroxostat improves vascular endothelial function in patients with hyperuricemia.

1. INTRODUCTION

Hyperuricemia is associated with hypertension1, 2 and an increased cardiovascular (CV) risk,3, 4, 5 although this remains to be established.6 Vascular endothelial dysfunction caused by the direct adverse effects of excessive uric acid7, 8 and oxidative stress induced by the reactive oxygen species (ROS) resulting from uric acid production9 is thought to explain this increased risk. Xanthine oxidoreductase (XOR) catalyzes serum uric acid (SUA) synthesis. Xanthine oxidase (XO), a type of XOR, is secreted from the internal organs into the bloodstream and binds strongly to the vascular endothelium and catalyzes SUA formation.10 Therefore, XOR inhibition contributes to normalizing vascular endothelial function, although it remains unclear whether this improvement results from lowered SUA levels or suppressed ROS production.

At present, three XOR inhibitors are used clinically in Japan, namely, allopurinol, febuxostat, and topiroxostat. Topiroxostat is a novel XOR inhibitor approved for the treatment of gout and hyperuricemia in Japan. In a previous study, it showed tolerability and non‐inferior SUA‐lowering effect to allopurinol in hyperuricemic patients with or without gout.11 Furthermore, topiroxostat and febuxostat showed similar renal protective and anti‐inflammatory effects after 6‐month treatment in hyperuricemic patients with cardiovascular disease.12 In recent years, the organ protective effect of these novel XOR inhibitors has been suggested, which has gained considerable attention.

Improvements in vascular endothelial function have already been shown with allopurinol and febuxostat therapy in some clinical research. Flow‐mediated dilation (FMD) is used as a measure of vascular endothelial function. Research has shown that patients with hyperuricemia and CV risks have lower ∆FMD values (percentage change in brachial artery diameter at peak from baseline) than people with normal SUA levels, and that allopurinol therapy improved ∆FMD in the patients with hyperuricemia after 3 months.13 Furthermore, allopurinol therapy has been shown to improve ∆FMD in patients with hyperuricemia and chronic kidney disease (CKD)14 as well as in patients without other risk factors for endothelial dysfunction, besides hyperuricemia.15 Febuxostat has also been reported to have a similar effect on ∆FMD in patients with hyperuricemia undergoing hemodialysis.16

Although topiroxostat lacks the direct evidence of improvements in endothelial function, a phase III clinical trial showed that it significantly lowered the urine albumin‐creatinine ratio compared with placebo in patients with stage G3 CKD,17 and another clinical trial demonstrated anti‐albuminuric effects of topiroxostat in patients with diabetic nephropathy.18 This evidence suggests an improvement effect of topiroxostat on vascular endothelial function in patients with hyperuricemia.

Given these report, we aimed to perform a retrospective cohort study to determine whether topiroxostat has beneficial effects on vascular function, including endothelial function, in patients with hyperuricemia.

2. METHODS

2.1. Study design and subjects

This was a retrospective, single‐arm, observational cohort study to evaluate the effects of topiroxostat on vascular function in hyperuricemic patients treated at Seichokai Medical Corporation Washiya Hospital. We selected patients who had already received topiroxostat and extracted only the necessary information from their medical charts. The study followed the “Ethical Guidelines for Medical and Health Research Involving Human Subjects,”19 and the study protocol was approved by the external Hattori Clinic Ethics Review Committee.

The inclusion criteria were as follows: (a) patients who began topiroxostat treatment between September 4, 2013 (when topiroxostat became available) and June 9, 2017 (the start day of the study); (b) patients who were diagnosed with at least one of the following: hypertension, diabetes mellitus, metabolic syndrome, hyperlipidemia, CKD, coronary artery disease, heart failure, aortic disease, or peripheral arterial disease; and (c) patients who provided written informed consent to participate in the study. No exclusion criteria were applied.

2.2. End points

The primary end point was the change in ∆FMD (percentage change in brachial artery diameter at peak from baseline) evaluated by ultrasonography from baseline to approximately week 8 of topiroxostat therapy. The secondary end points were the differences in the other CV and laboratory parameters at approximately week 8 of topiroxostat therapy.

2.3. Data collection

Data were extracted from baseline (182‐0 days prior to the estimated start of topiroxostat therapy) and approximately at 8 weeks of follow‐up (28‐119 days after the estimated start of topiroxostat therapy) for all patients. Only data derived from routine medical practice were collected and utilized. Observation was censored from the dates when topiroxostat was discontinued or another SUA‐lowering drug was added, as well as from the date of the last visit if patients had no medical records beyond 6 months. We primarily extracted the ΔFMD, but we also extracted changes in CV and serum parameters to evaluate the effects of topiroxostat therapy on vascular function. In addition, demographic data at baseline, duration of hyperuricemia since diagnosis, medical history, concomitant drug use, and concomitant non‐drug therapy were also extracted from the medical records. The presence or absence of metabolic syndrome was concluded based on the Japanese diagnostic criteria for metabolic syndrome.20

ΔFMD was measured by a system of ultrasonography equipped with software, as previously described (UNEX EF 18G; Unex Co. Ltd.).21, 22, 23 ΔFMD was automatically calculated using the following formula: [(diameter at peak − diameter at rest)/diameter at rest] × 100.

The other extracted parameters were the brachial artery diameter (at rest), ankle‐brachial index (ABI), brachial‐ankle pulse wave velocity (baPWV), augmentation index (AIx), and pressure wave reflection parameters (forward pulse height, reflected pulse height, and reflection index). We also extracted the systolic/diastolic brachial blood pressure (bBP), systolic/diastolic central blood pressure (cBP), 24‐hour blood pressure (mean, daytime, nighttime, and early morning), and blood pressure variation, as well as the central pulse pressure (cPP), the augmentation pressure, and the heart rate. Serum values were extracted for N‐terminal pro‐brain natriuretic peptide (NT‐proBNP), SUA, total cholesterol, low‐density lipoprotein cholesterol, high‐density lipoprotein cholesterol, triglyceride, serum creatinine, fasting blood glucose, and hemoglobin A1c.

2.4. Statistical analysis

SAS version 9.3 (SAS Institute) was used to analyze the data, and the significance level was set at a P‐value of <.05. Corrections for multiple testing were not performed in this study. When data were available at both evaluation points (baseline and follow‐up), the Shapiro‐Wilk test was performed using the data obtained at approximately 8 weeks after starting therapy. When the P‐value of Shapiro‐Wilk test was ≥.05, the paired t‐test was conducted. In other cases, Wilcoxon's signed‐rank test was conducted.

For exploratory analysis, correlation between percentage changes in the ∆FMD and the SUA level, as well as the correlation between change in ∆FMD and SUA levels at approximately 8 weeks after starting therapy, was analyzed using Pearson's product‐moment correlation coefficient. The P‐values and the coefficient of determination (R ²) were also calculated. Furthermore, for the ∆FMD, the means and 95% confidence intervals (CIs) of the differences between the two evaluation points were calculated in subgroups stratified by history of SUA‐lowering drug use, SUA levels at approximately 8 weeks after starting therapy (cut‐off values: 5.0, 6.0, and 7.0 mg/dL), and concomitant anticoagulant or antiplatelet drug use. The 95% CI for the mean was calculated using a t distribution.

3. RESULTS

In total, 23 patients (21 males and two females) with a mean age of 69.2 years met the inclusion criteria (Table 1). All participants were diagnosed with asymptomatic hyperuricemia and none had a history of gouty arthritis or tophus. The most common complication was hypertension (20 patients, 87.0%), with most patients using antihypertensives (21 patients, 91.3%). The second most common complication was hyperlipidemia (11 patients, 47.8%), with most of these (9 patients, 39.1% overall) receiving lipid‐lowering drugs. There were no documented changes in the dosages of these therapies during the study period. No patient had a history of non‐drug treatment, such as chronic dialysis, kidney transplantation, revascularization, or artificial blood vessel replacement (data not shown).

Table 1.

Baseline characteristics of the study subjects (N = 23)

Variables	n (%) or Mean ± SD
Sex (Male/Female)	21/2
Age (y)	69.2 ± 9.8
BMI (kg/m2)	26.3 ± 5.1
<18.5	1 (4.3%)
18.5 to <25	10 (43.5%)
25 to <30	8 (34.8%)
30 to <35	2 (8.7%)
≥35	2 (8.7%)
Drinking habit
Yes	7 (30.4%)
Yes (history)	0 (0%)
No	2 (8.7%)
Missing	14 (60.9%)
Smoking habit
Yes	3 (13.0%)
Yes (history)	2 (8.7%)
No	5 (21.7%)
Missing	13 (56.5%)
Duration of hyperuricemia (y)	2.3 ± 4.4
Complications
Gouty arthritis	0 (0%)
Gouty tophus	0 (0%)
Hypertension	20 (87.0%)
Diabetes mellitus	4 (17.4%)
Metabolic syndrome	0 (0%)
Hyperlipidemia	11 (47.8%)
CKD	0 (0%)
Coronary artery disease	3 (13.0%)
Heart failure	2 (8.7%)
Aortic disease	3 (13.0%)
PAD	2 (8.7%)
Concomitant drug usage
Antihypertensive drugs	21 (91.3%)
Coronary vasodilators	0 (0%)
Antiplatelet drugs	4 (17.4%)
Anticoagulant drugs	5 (21.7%)
Antidiabetic drugs	4 (17.4%)
Lipid‐lowering drugs	9 (39.1%)
History of other uric acid‐lowering drugs	5 (21.7%)
Serum uric acid (mg/dL)	7.36 ± 1.42

Abbreviations: BMI, body mass index; CKD, chronic kidney disease; PAD, peripheral arterial disease; SD, standard deviation.

The final daily topiroxostat dose during the observation period was 40 mg (n = 2) and 80 mg (n = 21), which are the approved doses. Five patients had received other SUA‐lowering drugs before topiroxostat. These were allopurinol (n = 4) and febuxostat (n = 1). The mean (SD) SUA level at baseline was 7.36 ± 1.42 mg/dL (n = 23), with levels of 7.72 ± 0.98 mg/dL in the treatment‐naïve patients and levels of 6.08 ± 2.11 mg/dL in the patients who switched from another SUA‐lowering drug.

Clinical test values, including SUA and CV parameters at baseline and approximately 8 weeks after starting therapy, are summarized in Table 2. The primary end point of change in ∆FMD from baseline to approximately week 8 was 1.00% ± 2.21% and was statistically significant (P = .045). The statistical power to detect the observed difference in ∆FMD with a significance level of 5% (two‐tailed test) was 53.1%.

Table 2.

Comparison of test results between baseline and approximately 8 wk after treatment with topiroxostat

Test item	N	Baseline	8 wk after	P‐value
Serum uric acid (mg/dL)	22	7.31 ± 1.43	5.44 ± 1.11	<.001*
∆FMD (%)	22	4.53 ± 2.09	5.54 ± 3.08	.045*
Brachial artery diameter (at rest) (mm)	22	4.23 ± 0.68	4.05 ± 0.58	.001*
Systolic bBP (mm Hg)	19	122.8 ± 13.0	122.4 ± 14.1	.905*
Diastolic bBP (mm Hg)	19	74.7 ± 9.7	72.6 ± 10.5	.203*
Heart rate (bpm)	19	65.5 ± 10.2	65.9 ± 13.3	.758†
baPWV (cm/s)	21	1740 ± 426	1757 ± 332	.690*
ABI (−)	22	1.098 ± 0.108	1.109 ± 0.111	.540*
Systolic cBP (mm Hg)	19	114.8 ± 12.4	114.8 ± 14.5	1.000*
Diastolic cBP (mm Hg)	19	75.3 ± 10.6	73.6 ± 10.6	.290*
cPP (mm Hg)	19	39.6 ± 10.1	41.3 ± 11.9	.521*
Augmentation pressure (mm Hg)	19	13.2 ± 5.5	13.1 ± 5.8	.939*
AIx (%)	19	32.6 ± 11.0	30.1 ± 7.0	.347†
AIx at heart rate of 75 bpm (%)	18	26.4 ± 6.7	25.4 ± 6.6	.410*
Forward pulse height (mm Hg)	15	25.9 ± 5.9	27.5 ± 5.9	.426*
Reflected pulse height (mm Hg)	15	18.6 ± 4.3	19.3 ± 5.2	.661*
Reflection index (%)	15	74.5 ± 12.9	68.6 ± 10.1	.118*
Total cholesterol (mg/dL)	22	186.7 ± 34.1	186.1 ± 34.3	.855*
LDL cholesterol (mg/dL)	22	105.6 ± 27.5	105.4 ± 27.6	.539†
HDL cholesterol (mg/dL)	22	55.3 ± 15.1	56.5 ± 16.2	.214*
Triglyceride (mg/dL)	22	126.6 ± 53.6	118.9 ± 57.1	.289*
Serum creatinine (mg/dL)	22	0.933 ± 0.176	0.918 ± 0.175	.262*
Fasting blood glucose (mg/dL)	22	103.4 ± 20.7	107.3 ± 25.6	.211*
HbA1c (NGSP) (%)	22	5.92 ± 0.85	5.89 ± 0.95	.193†
NT‐proBNP (pg/mL)	22	153.0 ± 249.9	235.5 ± 459.4	.325†

Values are expressed as mean ± standard deviation.

Abbreviations: ABI, ankle‐brachial index; AIx, augmentation index; baPWV, brachial‐ankle pulse wave velocity; bBP, brachial blood pressure; cBP, central blood pressure; cPP, central pulse pressure; FMD, flow‐mediated dilation; HbA1c, hemoglobin A1c; HDL, high‐density lipoprotein; LDL, low‐density lipoprotein; NGSP, national glycohemoglobin standardization program.

^* p‐value by paired t‐test

^† p‐value by Wilcoxon signed‐rank test.

Among the secondary end points, the SUA level and the resting brachial artery diameter decreased significantly. However, no significant changes were observed in the other CV parameters. The 24‐hour blood pressure results were not available for most patients, and hence are not reported.

No significant correlation was observed between the percentage increase in ∆FMD and the percentage decrease in SUA levels after approximately 8 weeks of topiroxostat therapy (R ² = 0.0108, P = .645) (Figure 1). Similarly, no significant correlation was observed between the increase in ∆FMD and SUA levels at approximately 8 weeks after starting topiroxostat therapy (R ² = 0.1286, P = .101) (Figure 2).

Figure 1.

Correlation between percentage increase in ∆ flow‐mediated dilation and percentage decrease in serum uric acid levels from baseline to approximately week 8

Figure 2.

Correlation between increases in ∆ flow‐mediated dilation from baseline to approximately week 8 and serum uric acid levels at approximately week 8

By contrast, subgroup analyses revealed that ∆FMD was significantly increased in groups with SUA levels of ≤6.0 or ≤7.0 mg/dL after approximately 8 weeks of topiroxostat therapy (Table 3). A numerical increase in the ∆FMD was also observed in the group with SUA levels ≤5.0 mg/dL, but the increase was not significant. The other factors assessed in the subgroup analyses were not relevant to the increase in ∆FMD.

Table 3.

Changes in ∆FMD from baseline to approximately week 8 in the subgroups

Subgroup	N	∆FMD (%)			P‐value
		Mean	95%CI	Median
All subjects (main analysis)	22	1.00	0.03‐1.98	1.00	.045
History of SUA‐lowering drug usage (−)	17	0.77	−0.20‐1.74	0.90	.110
History of SUA‐lowering drug usage (+)	5	1.80	−2.20‐5.80	3.00	.280
SUA after approximately 8 wk (≤5.0 mg/dL)	7	1.29	−1.13‐3.70	1.40	.241
SUA after approximately 8 wk (>5.0 mg/dL)	15	0.87	−0.28‐2.02	0.90	.125
SUA after approximately 8 wk (≤6.0 mg/dL)	16	1.36	0.11‐2.60	1.40	.035
SUA after approximately 8 wk (>6.0 mg/dL)	6	0.07	−1.62‐1.75	0.60	.923
SUA after approximately 8 wk (≤7.0 mg/dL)	21	1.15	0.17‐2.13	1.10	.024
SUA after approximately 8 wk (>7.0 mg/dL)	1	−2.00	‐	−2.00	‐
Concomitant use of A/C or A/P (−)	13	1.10	−0.42‐2.62	1.40	.140
Concomitant use of A/C or A/P (+)	9	0.87	−0.52‐2.26	0.80	.188

P‐value by paired t‐test.

Abbreviations: A/C, anticoagulant; A/P, antiplatelet; CI, confidence interval; FMD, flow‐mediated dilation; SUA, serum uric acid.

4. DISCUSSION

This is the first study to suggest the beneficial effects of topiroxostat therapy on vascular endothelial function evaluated by ultrasonography. We found that approximately 8 weeks of therapy increased the brachial artery ∆FMD by approximately 1.0%, from 4.53% ± 2.09% to 5.54% ± 3.08%, in patients with hyperuricemia and CV risks.

Assessment of brachial artery ∆FMD was proposed by Celermajer et al24 and is recommended as a noninvasive diagnostic measure for vascular endothelial function in clinical guidelines.25, 26 A ∆FMD of ≤4.5% measured in the acute phase of ischemic stroke has been shown to be an independent predictor of new‐onset vascular events (hazard ratio = 3.48; 95% CI, 1.26‐9.63) in patients with a history of acute ischemic stroke.27 A meta‐analysis of 14 studies also indicated that a 1% increase in the forearm ∆FMD and brachial artery ∆FMD decreased the risk of new‐onset CV events by 9% (95% CI, 4%‐13%) and 17% (95% CI: 12%‐22%), respectively.28 Therefore, we consider that the improvement in ∆FMD after topiroxostat therapy in this study was clinically relevant.

The effect of lowering SUA levels on FMD may be altered by basal ∆FMD. Maruhashi et al29 reported that ∆FMD significantly decreased with an increase in the SUA level, particularly in postmenopausal women (mean ∆FMD: 5.5%), but not in premenopausal women (mean ∆FMD: 7.8%). They reported the possibility of treatment of hyperuricemia in postmenopausal women improving endothelial dysfunction. In addition, a recent observational study reported that there was a modest but significant correlation between basal ∆FMD and SUA level in the overall study subjects with mean basal ∆FMD of 4.9%, although no significant association was observed in the analysis limited to male subjects.30 Remarkably, similar to the present study (mean basal ∆FMD: 4.5%), the mean basal ∆FMD in most previous studies, which showed improvements in vascular endothelial function with the use of XOR inhibitors, was lower than 6.0% ,13, 14, 16 which is the recommended reference value of ∆FMD.31 These evidences suggest that hyperuricemic patients with lower ∆FMD will be more susceptible to the beneficial effect of SUA‐lowering drugs on vascular endothelial function.

We also showed that the brachial artery diameter at rest decreased significantly by approximately 8 weeks of topiroxostat therapy. Some studies have reported that the brachial artery diameter is an independent predictor of CV events.32, 33 In addition, a study using a rat carotid artery ligation model reported that 3‐week treatment with XOR inhibitor inhibited neointimal hyperplasia.34 Therefore, it is possible that improved vascular endothelial function during topiroxostat therapy led to suppressed arterial remodeling and a decrease in the brachial artery diameter. However, further clinical studies are warranted to clarify the effect of topiroxostat on vascular remodeling considering the limited clinical evidence on this matter.

Although a recent observational study reported that hyperuricemia has significant independent longitudinal association with elevation in baPWV,35 no significant changes were observed in the parameters for arterial stiffness such as baPWV, ABI, and AIx after topiroxostat therapy in the present study. Remarkably, a recent meta‐analysis suggested some favorable effects of allopurinol treatment on improving ∆FMD36 and AIx but not on PWV.37 The changes in AIx following allopurinol treatment in RCTs included in the meta‐analysis were from −5.54% to +2.2%, almost similar to that observed in this study (−2.5%). Although the reason for this discrepancy among the parameters of vascular functions remains unclear, one possible explanation is that XOR inhibitors improve vascular function at an earlier stage of arteriosclerosis and have a greater influence on microvascular functions than on central macrovascular functions.37 Another possible explanation is that it takes relatively long time for the improvement of PWV following the improvement of endothelial function. In fact, Tomiyama et al38 reported that endothelial dysfunction is associated with the progression of arterial stiffness (assessed by baPWV) after 3‐year follow‐up in hypertensive patients. We suggest that studies with long‐term follow‐ups are needed to clarify the effects of topiroxostat on the parameters of arterial stiffness.

Topiroxostat therapy was associated with a significant decrease in the SUA levels from 7.31 ± 1.43 mg/dL at baseline to 5.44 ± 1.11 mg/dL at follow‐up. Mercuro et al13 demonstrated that a clear correlation existed between the percentage increase in ∆FMD and the percentage decrease in SUA levels in patients receiving allopurinol for hyperuricemia. Although we failed to show a significant correlation between the percentage increase in ∆FMD and percentage decrease in SUA levels, subgroup analyses by SUA level at approximately week 8 indicated that lower SUA levels at this point may be associated with greater improvements in vascular endothelial function. These results suggest that low SUA levels at approximately 8 weeks after starting topiroxostat therapy were associated with the increased ∆FMD. In other words, it suggests that, rather than the amount of change, lowering SUA levels less than a cut‐off point is crucial. Prior research showed no improvement in endothelial function after 3 months of allopurinol therapy in normouricemic patients with CV risks, but did show a slight decrease in SUA levels from 5.4 to 4.9 mg/dL.13 Thus, there appears to be no advantage in lowering SUA levels among patients with SUA levels below normal. The optimal SUA range for organ protection in patients with hyperuricemia should be addressed in future research.

The present study indicated that topiroxostat has the potential to improve vascular endothelial function in patients with hyperuricemia and CV risks. Patients with hypertension and higher SUA levels have higher albuminuria.39 Furthermore, albuminuria and vascular endothelial dysfunction are closely related in the renal microcirculation.40 A phase III clinical trial in patients with hyperuricemia and stage G3 CKD reported that topiroxostat lowered the urinary albumin‐to‐creatinine ratio by approximately 33%, and the difference was significant compared with placebo after 22 weeks of therapy.17 This result would be associated with topiroxostat improving endothelial function.

Recently, we reported that an integrated FMD response, calculated as the area under the dilation curve over a 120‐seconds dilation period (FMD‐AUC₁₂₀), could predict CV events over 5 years in elderly patients with CV risk factors (ie, hypertension, dyslipidemia, diabetes mellitus, or smoking).41 Since no study on FMD‐AUC₁₂₀ has been done in patients with hyperuricemia, it is important to determine which of it or ∆FMD is the best predictor of CV events in these patients.

A preclinical study using an ischemia‐reperfusion model indicated that the CV protective effect of topiroxostat was through suppression of ROS generation.42 Indeed, it remains controversial whether the improved vascular endothelial function during therapy with XOR inhibitors is due to their SUA‐lowering effects or the suppression of ROS production. Benzbromarone, a uricosuric agent that does not inhibit XOR, was also reported to improve vascular endothelial function43 and to reduce the incidence of coronary artery disease as effectively as allopurinol in a retrospective study.44 Furthermore, a recent study using a claims database reported on the incidence rates of CV events among patients with gout receiving allopurinol or probenecid (a uricosuric agent).45 Of note, the incidence of CV events was lower in patients who received probenecid than in those who received allopurinol; that said, although this was a nonrandomized study that was limited by concerns around medication adherence. Pharmacological experiments have also shown that uric acid itself induces oxidative stress and promotes apoptosis.46, 47 Another in vitro study showed that uric acid reduced NO production in the endothelial cells, and it could be normalized by SUA‐lowering drugs.48 Therefore, it is plausible that SUA reduction may be one of the major mechanisms of action of topiroxostat in improving vascular endothelial function.

5. LIMITATIONS

This study has several limitations that must be taken into consideration when interpreting the results. First, limitations derived from the retrospective study design were required to be considered. The presence of systematic errors due to missing data (measurement bias) cannot be denied, and we were unable to confirm that the examination results were obtained under the same conditions and evaluate the reproducibility of examinations, such as ∆FMD measurement. However, we consider the measurement bias to be minimal because this was a single‐site study using standardized methods of examination. Additionally, because we allowed extended date ranges for data collection at baseline and follow‐up, the duration of topiroxostat therapy was not consistent. Moreover, five patients were taking other SUA‐lowering drugs prior to topiroxostat therapy, although no significant effect was observed in the subgroup analysis. The second limitation is that the study results might be influenced by the progression of complications (natural course) and the concomitant use of other drugs because of the single‐arm, non‐comparative design, without adjustment for confounders. In addition, because the subjects of the present study were heterogeneous in terms of CV risk factors, the study results might be influenced by their effect if there are interactions between the effect of topiroxostat on ∆FMD and these CV risk factors. However, the short‐term effects of any concomitant drug on ∆FMD were probably minimal because all dosages remained stable and there were no significant differences in blood pressure, cholesterol, triglyceride and glucose levels during the study, although the long‐term effects of the drugs for chronic diseases, such as statins and antihypertensives, cannot be denied. Single‐arm and non‐blinded studies are generally susceptible to the Hawthorne effect,49 a phenomenon wherein a study subject's behavior and/or study outcomes are altered by the subject's awareness of being under observation. However, we consider the Hawthorne effect was minimal because our study was retrospective in nature and used only the data derived from routine medical practice and those collected retroactively. Lastly, our sample size was slightly small considering the heterogeneity of the study subjects. In fact, as the statistical power to detect observed difference in ∆FMD was not very high (53.1%), larger sample size may be required to confirm the reproducibility of the results in future studies.

6. CONCLUSIONS

We conclude that topiroxostat therapy may improve vascular endothelial function in patients with hyperuricemia. However, it remains necessary to clarify the mechanism of action for this effect and to explore the patient types who will benefit most from therapy. In addition, considering the limitations owing to retrospective study design, prospective studies on a larger homogenous population with a control group are needed to clarify the exact role of topiroxostat on vascular endothelial function as the next step.

CONFLICT OF INTEREST

SH, DS, and YF are full‐time employees at Pfizer Japan Inc. KK has received research grants from A&D Co., Bayer Yakuhin, Boehringer Ingelheim, Daiichi Sankyo, EA Pharma, Fukuda Denshi, Medtronic, Mitsubishi Tanabe Pharma Corporation, Mochida Pharmaceutical Co., Omron Healthcare, Otsuka, Pfizer, Takeda, and Teijin Pharma; and honoraria from Daiichi Sankyo, Omron Healthcare, and Takeda. NT has no conflicts of interest to declare.

AUTHOR CONTRIBUTIONS

All authors were involved in conceptualizing and designing the study, interpreting the results, and drafting and revising the manuscript. KK and NT developed the hypothesis, idea, and protocol. DS and YF also developed the protocol. SH acquired, managed, and analyzed the data. All authors read and approved the final version of the manuscript.

Higa S, Shima D, Tomitani N, Fujimoto Y, Kario K. The effects of topiroxostat on vascular function in patients with hyperuricemia. J Clin Hypertens. 2019;21:1713–1720. 10.1111/jch.13707