Circ J 2017; 81: 1707 – 1712 ORIGINAL ARTICLE doi: 10.1253/circj.CJ-17-0438 Preventive Medicine

Cross-Over Trial of and for With Cardiovascular Disease (TROFEO Trial)

Akira Sezai, MD, PhD; Kazuaki Obata; Keisuke Abe; Sakie Kanno; Hisakuni Sekino, MD, PhD

Background: We previously reported that febuxostat was more effective for hyperuricemia than . The efficacy, however, of topiroxostat (a novel oxidase reductase inhibitor similar to febuxostat), for hyperuricemia is unknown.

Methods and Results: Patients with cardiovascular disease and hyperuricemia, in whom serum (s-UA) was controlled at ≤6 mg/dL, were eligible for enrollment. Fifty-five patients were randomized to receive either febuxostat or topiroxostat for 6 months and were switched to the other drug for the following 6 months. The primary endpoint was s-UA. Secondary endpoints included serum creatinine, estimated glomerular filtration rate, urinary albumin, cystatin-C, oxidized low-density lipoprotein, eicosapentaenoic acid/ arachidonic acid ratio, lipid biomarkers, high-sensitivity C-reactive protein and B-type natriuretic protein. Although s-UA level was similar for both drugs, significantly more patients required dose escalation during treatment with topiroxostat. There were no differ- ences in renal function, inflammatory and lipid markers between the 2 drugs. A biomarker of oxidative stress was significantly lower after 3 months of febuxostat compared with topiroxostat.

Conclusions: Febuxostat causes more marked and more rapid reduction of s-UA than topiroxostat. With regard to the antioxidant effect, febuxostat was superior to topiroxostat after 3 months. The renal protective and anti-inflammatory effects of both drugs were also similar after 6 months of treatment. Thus, both of these agents were similarly effective for hyperuricemia in patients with cardio- vascular disease.

Key Words: Febuxostat; Hyperuricemia; Topiroxostat

yperuricemia was recently reported to be associ- to reduce UA more potently than allopurinol,7 while ated with hypertension, cardiovascular disease, topiroxostat has a similar effect to allopurinol.8 No study, H and chronic kidney disease (CKD).1,2 Allopurinol however, has compared febuxostat with topiroxostat. has long been regarded as a first-line drug for the treatment Accordingly, a clinical study was conducted to compare of hyperuricemia, but adverse reactions such as renal dys- febuxostat with topiroxostat and assess potential differ- function, Stevens-Johnson syndrome, and hypersensitivity ences in efficacy (TopiROxostat and FEbuxostat in a ran- vasculitis have been reported with allopurinol, and it is not domized, Open-label, cross-over trial for hyperuricemia sufficiently effective in some cases.3,4 In a comparative trial with cardiovascular disease (TROFEO trial). of febuxostat vs. allopurinol, we reported that febuxostat reduced serum uric acid (s-UA) earlier than allopurinol, Methods had a stronger renoprotective effect than allopurinol, and also had superior antioxidant and anti-inflammatory Study Protocol effects.5,6 Unlike allopurinol, febuxostat and topiroxostat The subjects were outpatients with cardiovascular dis- (a novel reductase [XOR] inhibitor) have ease and hyperuricemia in whom s-UA was controlled at a non- structure and are mainly metabolized in the ≤6 mg/dL by treatment with allopurinol or febuxostat. In liver. In addition, these agents are excreted in both the this study, patients were randomized by the envelop urine and feces. Accordingly, their impact on the kidney is method to receive treatment with either febuxostat (Teijin minimal, and dose adjustment depending on renal function Pharma, Tokyo, Japan) or topiroxostat (Sanwa Kagaku is not required, in contrast to allopurinol. These 2 drugs Kenkyusho, Aichi, Japan and Fujiyakuhin, Saitama, have different dosing regimens (once daily for febuxostat Japan) for 6 months, after which they switched to the other vs. twice daily for topiroxostat) and febuxostat is reported for another 6 months. Baseline data were

Received April 24, 2017; accepted May 4, 2017; released online June 9, 2017 Time for primary review: 10 days Department of Cardiovascular Surgery, Nihon University School of Medicine, Tokyo (A.S.); Sekino Hospital, Tokyo (K.O., K.A., S.K., H.S.), Japan Registration: UMIN (http://www.umin.ac.jp/), Study ID: UMIN000014771 Mailing address: Akira Sezai, MD, PhD, Department of Cardiovascular Surgery, Nihon University School of Medicine, 30-1 Oyaguchi-kamimachi, Itabashi-ku, Tokyo 173-8610, Japan. E-mail: [email protected] ISSN-1346-9843 All rights are reserved to the Japanese Circulation Society. For permissions, please e-mail: [email protected]

Circulation Journal Vol.81, November 2017 1708 SEZAI A et al.

Table 1. Patient Characteristics for allopurinol or one-quarter of that for topiroxostat. Topiroxostat was given twice daily (after breakfast and Characteristic Data dinner) at a dose two-fifths of that for allopurinol and n 55 4-fold that for febuxostat. During the study period, s-UA Age (years) 67.9±9.0 was measured at monthly intervals. If it was ≥6.0 mg/dL, (39–79) the dose of febuxostat was increased by 10 mg/day or that 43:12 Gender (M:F) of topiroxostat was increased by 40 mg/day. The dose was Basic disease increased up to a maximum of 60 mg/day for febuxostat or Ischemic heart disease 20 (36) 160 mg/day for topiroxostat. Exclusion criteria were (1) Valvular disease 23 (42) renal dysfunction with an estimated glomerular filtration Aortic disease 10 (18) rate (eGFR) ≤20 mL/min/1.73 m2; (2) hepatic dysfunction Others 2 (4) (aspartate aminotransferase [AST] >39 U/L or alanine Risk factors aminotransferase [ALT] >44 U/L); (3) treatment with mer- Diabetes mellitus 19 (35) captopurine hydrate or azathiopurine; (4) ; (5) Hypertension 45 (82) other reasons that made patients unsuitable for this study Dyslipidemia 40 (73) as judged by the attending physician. This study was con- ducted at Sekino Hospital, a logistical support hospital of Chronic kidney disease 39 (71) Nihon University Itabashi Hospital. The details of the Cerebrovascular disease 4 (7) study were explained to the patients and informed consent Obesity 7 (13) was obtained. Approval of the institutional review board Smoking 18 (33) was also obtained and the study was registered with Medication the Hospital Medical Information Network (study ID: ARB 29 (53) UMIN000014771). ACEI 4 (7) Renin inhibitor 6 (11) Endpoints Aldosterone blocker 29 (53) The primary endpoint was s-UA level after treatment. The antagonist 23 (42) secondary endpoints were as follows: serum creatinine β-blocker 38 (69) (s-Cr), eGFR, urinary albumin, cystatin-C, oxidized low- density lipoprotein (O-LDL), eicosapentaenoic acid/ara- Statin 40 (73) chidonic acid (EPA/AA) ratio, total cholesterol (T-cho), Furosemide 25 (45) triglycerides (TG), low-density lipoprotein (LDL), high- Febuxostat 47 (85) density lipoprotein (HDL), remnant-like particle-cholesterol Allopurinol 8 (15) (RLP-cho), high-sensitivity C-reactive protein (hs-CRP), Data given as n (%) or mean ± SEM (range). ACEI, angiotensin- B-type natriuretic peptide (BNP), and adverse reactions. converting inhibitor; ARB, angiotensin II blocker. s-UA, s-Cr, eGFR, T-cho, TG, LDL, HDL, and LDL/HDL (L/H) were measured before the start of treatment as well as after every month of treatment, while urinary albumin, cystatin-C, O-LDL, EPA/AA ratio, and BNP were mea- sured before treatment and after 3 and 6 months of treat- ment. Adverse reactions were classified as acute attacks of , skin reactions, renal dysfunction (increase of s-Cr by ≥50%), hepatic dysfunction (increase of AST/ALT by ≥50%), gastrointestinal symptoms, and allergic reactions. Management of the reactions (discontinuation of the test drug, etc.) was decided by the attending physician.

Statistical Analysis Measured values are expressed as mean ± SEM. Two-way analysis of variance (ANOVA) was used to compare param- eters between the febuxostat and topiroxostat groups, and P<0.05 was considered statistically significant.

Results Patients Fifty-five patients were enrolled in this trial and their Figure 1. Study flowchart. baseline characteristics are listed in Table 1. There were no drop-outs and all patients completed the 1-year study period without complications. Although there was a differ- ence between the 2 groups with regard to the obtained prior to switching to febuxostat or topiroxostat used for hyperuricemia prior to the study, there were no and monitoring was continued for 6 months after switch- other differences in patient characteristics. Before study ing. initiation, 22 patients in the febuxostat group were receiv- At the time of switching medication, febuxostat was ing febuxostat and 4 were taking allopurinol, while 25 given once daily after breakfast at a dose one-tenth of that patients in the topiroxostat group were receiving febuxo-

Circulation Journal Vol.81, November 2017 TROFEO Trial 1709

Table 2. Change in Renal Function, Lipid Parameters and hs-CRP Before 1 month 2 months 3 months 4 months 5 months 6 months treatment s-Cr (mg/dL) Febuxostat 1.28±0.58 1.24±0.58 1.19±0.59 1,22±0.58 1.21±0.59 1.22±0.65 1.20±0.54 Topiroxostat 1.19±0.53 1.21±0.55 1.21±0.47 1.22±0.56 1.23±0.54 1.24±0.64 1.30±0.68 eGFR (ml/min/1.73 m2) Febuxostat 45.6±2.1 47.2±2.0 47.4±1.9 47.6±2.0 47.7±2.2 47.4±2.0 45.3±2.1 Topiroxostat 49.0±2.0 47.7±2.0 47.9±2.0 48.1±2.1 47.3±2.1 48.3±2.1 48.2±2.0 T-cho (mg/dL) Febuxostat 163.9±4.0 162.9±4.0 164.0±4.0 165.0±4.3 167.4±4.0 164.2±3.9 160.2±3.7 Topiroxostat 159.8±4.0 160.0±3.9 160.1±4.0 163.1±4.0 166.4±4.1 163.2±4.1 161.5±4.3 TG (mg/dL) Febuxostat 134.7±10.4 133.0±11.0 124.1±7.6 135.9±10.2 123.2±7.3 121.0±7.0 121.3±8.7 Topiroxostat 127.7±8.5 119.4±7.5 117.5±7.1 120.9±7.4 114.7±6.3 113.9±7.6 113.9±7.6 L/H Febuxostat 1.62±0.90 1.62±0.84 1.62±0.84 1.66±0.92 1.65±0.10 1.55±0.86 1.57±0.85 Topiroxostat 1.63±0.81 1.60±0.80 1.60±0.82 1.57±0.82 1.58±0.81 1.56±0.80 1.54±0.84 hs-CRP (mg/dL) Febuxostat 0.18±0.03 0.25±0.05 0.27±0.06 0.24±0.06 0.23±0.06 0.35±0.12 0.24±0.05 Topiroxostat 0.19±0.04 0.33±0.15 0.23±0.04 0.18±0.03 0.25±0.07 0.28±0.06 0.21±0.04 eGFR, estimated glomerular filtration rate; hs-CRP, high-sensitivity C-reactive protein; L/H, high-density lipoprotein/low-density lipoprotein; s-Cr, serum creatinine; T-cho, total cholesterol; TG, triglyceride.

stat and 4 were taking allopurinol. There was no significant difference in prior therapy for hyperuricemia (P=1.00). It was confirmed that all patients in both groups took the assigned medications correctly.

Primary Endpoint There was no significant difference in s-UA between the 2 groups either before or after treatment. s-UA, however, did not exceed 6.0 mg/dL in the febuxostat group during the study period, but it exceeded this level in 9 patients from the topiroxostat group, with the number being sig- nificantly higher in the topiroxostat group (P=0.003; Figure 1). The dose of febuxostat was 16.5±9.7 mg at the initiation of treatment. No patient required dose escalation of febuxostat and the same dose was maintained for 6 months. The dose of topiroxostat was 59.6±33.5 mg at ini- tiation of treatment, while it was 61.5±33.3 mg, 64.4±35.6 mg, 68.0±38.2 mg, 69.5±38.7 mg, 70.9±40.5 mg and 70.9±40.5 mg after 1, 2, 3, 4, 5 and 6 months, respectively.

Secondary Endpoints s-Cr, eGFR There was no significant difference in s-Cr and eGFR between the 2 drugs either before or after treat- ment (Table 2; Figure 2). Urinary Albumin There was no difference in urinary albumin prior to treatment with the 2 drugs, and no sig- nificant difference was observed after 3 or 6 months of treatment (3 months, P=0.369; 6 months, P=0.359; Figure 2). Cystatin-C There was no difference in cystatin-C before treatment with the 2 drugs, as well as no significant differ- ence after 3 or 6 months of treatment (3 months, P=0.359; 6 months, P=0.300; Table 3). O-LDL There was no difference in O-LDL prior to treat- ment with the 2 drugs. O-LDL was significantly lower after Figure 2. Change in (Upper) serum uric acid (UA) and 3 months of febuxostat treatment compared with topirox- (Lower) number of patients with xanthine oxidase reductase ostat (P=0.030), but there was no significant difference (XOR) inhibitor add-on therapy (Lower). between the 2 drugs after 6 months of treatment (P=0.227;

Circulation Journal Vol.81, November 2017 1710 SEZAI A et al.

Table 3. Change in Cystatin-C, EPA/AA, RLP-cho, and BNP Before Cystatin-C 3 months 6 months treatment Cystatin-C (mg/dL) Febuxostat 1.41±0.07 1.34±0.07 1.34±0.07 Topiroxostat 1.39±0.06 1.43±0.07 1.46±0.07 EPA/AA Febuxostat 0.56±0.05 0.66±0.06 0.61±0.06 Topiroxostat 0.73±0.12 0.61±0.06 0.57±0.05 RLP-cho (mg/dL) Febuxostat 5.31±0.48 5.17±0.47 4.69±0.46 Topiroxostat 5.20±0.50 5.51±0.43 5.05±0.37 BNP (pg/mL) Febuxostat 107.7±22.8 96.3±21.0 98.5±23.5 Topiroxostat 91.7±22.8 102.6±22.9 120.9±26.4 BNP, B-type natriuretic peptide; EPA/AA, eicosapentaenoic acid/arachidonic acid; RLP-cho, remnant-like particle- cholesterol.

Figure 3. Change in estimated glomerular filtration rate (eGFR), urinary albumin, and oxidized low- density lipoprotein (O-LDL).

Figure 2). patients experienced adverse reactions and there were no EPA/AA There was no significant difference in EPA/AA attacks of gout while they were on treatment with either ratio between the 2 drugs either before or after treatment drug. Among the 28 patients taking warfarin, there were (Table 3). no bleeding events or strokes during the study period. Nine T-cho, TG, LDL, HDL, and L/H There were no significant patients (32%), however, had an increase in international differences in T-cho, TG, LDL, HDL, and L/H between normalized ratio of prothrombin time (PT-INR) by ≥150% the 2 drugs either before or after treatment (Table 2). after switching from febuxostat to topiroxostat, and the RLP-cho There was no significant difference in RLP-cho dose of warfarin was reduced in all 9 patients. Conversely, between the 2 drugs either before or after treatment (Table 3). 3 patients (11%) had a reduction of PT-INR by ≥150% hs-CRP There was no significant difference in hs-CRP after switching from topiroxostat to febuxostat. Prior to between the 2 drugs either before or after treatment (Table 2). the study, there was no significant difference in PT-INR BNP There was no significant difference in BNP between between patients using either drug. PT-INR was 2.70±0.18 the 2 drugs either before or after treatment (Table 3). at 1 month after switching to topiroxostat vs. 2.01±0.10 at Other Parameters During the study period, none of the 1 month after switching to febuxostat, and it was signifi-

Circulation Journal Vol.81, November 2017 TROFEO Trial 1711

Figure 4. Change in international nor- malized ratio of prothrombin time (PT- INR).

cantly higher while the patients were taking topiroxostat compared with untreated patients.12 In a study comparing (P=0.002; Figure 3). febuxostat and topiroxostat in mice, dose-dependent reduction of urinary albumin and plasma XOR by topirox- Discussion ostat was noted, while febuxostat did not show dose- dependent activity. During treatment with topiroxostat, In the present cross-over study, we found that dose escala- there was a significant correlation between the changes of tion was required significantly more often for topiroxostat urinary albumin and plasma XOR activity, suggesting that than febuxostat over a 6-month treatment period (P=0.003), topiroxostat might reduce urinary albumin due to inhibi- but there was no difference in s-UA between the 2 drugs tion of plasma XOR activity.13 Our previous study of after 6 months. Reduction of s-UA was more rapid with febuxostat vs. allopurinol demonstrated significant reduc- febuxostat than topiroxostat. O-LDL, a marker of oxida- tion of urinary albumin by febuxostat.5,6 In the present tive stress, was significantly lower after 3 months of febux- study, there was no statistical difference in the effect on ostat treatment compared with topiroxostat treatment, but urinary albumin between febuxostat and topiroxostat (3 not after 6 months. Also, both drugs were largely similar months, P=0.369; 6 months, P=0.359). Accordingly, it with respect to their renoprotective, antioxidant, and anti- seems that both topiroxostat and febuxostat reduce uri- inflammatory effects at 6 months. It has been reported that nary albumin effectively in the clinical setting. febuxostat reduces UA more effectively than allopurinol,7 Fukui et al performed a clinical study in which allopuri- while the reduction of UA by topiroxostat and allopurinol nol was switched to febuxostat, and they measured deriva- is reported to be similar.8 In our previous study, febuxostat tives of reactive oxygen metabolites and the biological not only reduced UA more rapidly than allopurinol, but antioxidant potential as markers of oxidative stress. Both also achieved significantly lower s-Cr, eGFR, urinary albu- markers were significantly reduced after switching to min, cystatin-C, O-LDL, and hs-CRP compared with allo- febuxostat.14 In the present study, we measured O-LDL as purinol.5,6 Given that the patient demographic profile an index of oxidative stress and found that it was signifi- differed between our previous study and the present study, cantly lower after 3 months of treatment with febuxostat head-to-head comparison is not appropriate. Like febux- vs. topiroxostat, but there was no difference after 6 months. ostat, however, topiroxostat may also have a stronger In our previous comparative study of allopurinol and renoprotective effect, greater antioxidant activity, and a febuxostat, O-LDL was significantly lower in patients stronger anti-inflammatory effect than allopurinol, sug- receiving febuxostat.5,6 There have been no reports on the gesting that further investigation of this drug is warranted influence of topiroxostat on oxidative stress. Given that (Figure 4). subject demographics were different between our previous Uric acid can damage the kidneys and an excessive study and the present study, direct comparison of the increase of s-UA leads to a risk of CKD.9 Allopurinol results is not appropriate. It seems likely, however, that reduces s-UA, thereby suppressing the progression of renal topiroxostat would be more effective at inhibiting oxida- disease.10 According to clinical research on febuxostat, tive stress than allopurinol, as would be expected due to greater reduction of s-UA is associated with better mainte- reduction of oxidative stress by inhibiting XOR. nance of renal function.11 In the present study, warfarin activity was potenti- Urinary albumin is significantly reduced by topiroxostat ated by ≥150% in 32% of patients receiving topiroxostat.

Circulation Journal Vol.81, November 2017 1712 SEZAI A et al.

Although there have been no previous reports about M, et al. Association of serum uric acid with blood pressure in potentiation of the effect of warfarin by topiroxostat, the Japanese men: Cross-sectional study in work-site group. Circ J present finding suggests that this agent should be used with 2011; 75: 2827 – 2832. 3. Fagugli RM, Gentile G, Ferrara G, Brugnano R. Acute renal caution in patients on warfarin. and hepatic failure associated with allopurinol treatment. Clin Nephrol 2008; 70: 523 – 526. 4. Halevy S, Ghislain PD, Mockenhaupt M, Fagot JP, Bouwes Conclusions Bavinck JN, Sidoroff A, et al. Allopurinol is the most common cause of Stevens-Johnson syndrome and toxic epidermal necrol- Febuxostat causes more marked and more rapid reduction ysis in Europe and Israel. J Am Acad Dermatol 2008; 58: 25 – 32. of UA than topiroxostat. With regard to the antioxidant 5. Sezai A, Soma M, Nakata K, Hata M, Yoshitake I, Wakui S, et effect of these drugs, febuxostat was superior to topiroxo- al. Comparison of febuxostat and allopurinol for hyperuricemia stat after 3 months but no difference was observed after 6 in cardiac surgery patients (NU-FLASH Trial). Circ J 2013; 77: 2043 – 2049. months. The renal protective and anti-inflammatory effects 6. Sezai A, Soma M, Nakata K, Osaka S, Ishii Y, Yaoita H, et al. of both drugs were also similar after 6 months of treat- Comparison of febuxostat and allopurinol for hyperuricemia in ment. Thus, both of these agents were similarly effective cardiac surgery patients with chronic kidney disease (NU-FLASH for hyperuricemia in patients with cardiovascular disease. trial for CKD). J Cardiol 2015; 66: 298 – 303. 7. Becker MA, Schumacher HR, Wortmann RL, MacDonald PA, Eustace D, Palo WA, et al. Febuxostat compared with allopuri- Study Limitations nol in patients with hyperuricemia and gout. N Engl J Med 2005; This study was not conducted in treatment-naïve patients, 353: 2450 – 2461. and many of the subjects had previously shown a response 8. Hosoya T, Ogawa Y, Hashimoto H, Ohashi T, Sakamoto R. Comparison of topiroxostat and allopurinol in Japanese hyper- to febuxostat. In addition, the sample size was fairly small. uricemic patients with or without gout: A phase 3, multicentre, To confirm the present findings, a larger population of randomized, double-blind, double-dummy, active-controlled, treatment-naïve patients with hyperuricemia should be parallel-group study. J Clin Pharm Ther 2016; 41: 290 – 297. investigated in a similar study in the future. The results 9. Obermayr RP, Temml C, Gutjahr G, Knechtelsdorfer M, should assist in elucidating the advantages and disadvan- Oberbauer R, Klauser-Braun R. Elevated uric acid increases the risk for kidney disease. J Am Soc Nephrol 2008; 19: 2407 – 2413. tages of each medication. 10. Goicoechea M, de Vinuesa SG, Verdalles U, Ruiz-Caro C, Ampuero J, Rincón A, et al. Effect of allopurinol in chronic Disclosures kidney disease progression and cardiovascular risk. Clin J Am Soc Nephrol 2010; 5: 1388 – 1393. The authors declare no conflict of interest. 11. Whelton A, Macdonald PA, Zhao L, Hunt B, Gunawardhana L. Renal function in gout: Long-term treatment effects of febuxo- Financial Support stat. J Clin Rheumatol 2011; 17: 7 – 13. A.S. received funding from Daiichi Sankyo and Teijin, and this study 12. Hosoya T, Ohno I, Nomura S, Hisatome I, Uchida S, Fujimori was also supported by a Research Grant from the Japanese Ministry S, et al. Effects of topiroxostat on the serum urate levels and of Education, Culture, Sports, Science and Technology (No.21591805). urinary albumin in hyperuricemic stage 3 chronic kid- A.S. has received lecture fees from Daiichi Sankyo. The other authors ney disease patients with or without gout. Clin Exp Nephrol 2014; declare no conflict of interest. 18: 876 – 884. 13. Nakamura T, Murase T, Nampei M, Morimoto N, Ashizawa N, Iwanaga T, et al. Effects of topiroxostat and febuxostat on uri- References nary albumin excretion and plasma xanthine oxidoreductase 1. Takae K, Nagata M, Hata J, Mukai N, Hirakawa Y, Yoshida activity in db/db mice. Eur J Phamacol 2016; 780: 224 – 231. D, et al. Serum uric acid as a risk factor for chronic kidney dis- 14. Fukui T, Murayama M, Yamauchi K, Yoshitaka S, Yasuda T, ease in a Japanese community: The Hisayama Study. Circ J 2016; Abe Y. Effects of febuxostat on oxidative stress. Clin Ther 2015; 80: 1857 – 1862. 37: 1396 – 1401. 2. Kansui Y, Ohtsubo T, Goto K, Sakata S, Ichishima K, Fukuhara

Circulation Journal Vol.81, November 2017