CI‐1

Introduction and Regulatory History

Beth‐Anne Knapp Vice President Global Regulatory Affairs, Marketed Products CI‐2 Agenda for Today

Introduction Beth-Anne Knapp Vice President Global Regulatory Affairs, Marketed Products Gout Disease Background Michael A. Becker, MD and Treatment Landscape Professor Emeritus University of Chicago CARES and Cardiovascular Safety of William B. White, MD Febuxostat Professor of Medicine, Cardiology Center University of Connecticut School of Medicine Efficacy Lhanoo Gunawardhana, MD, PhD Senior Medical Director Clinical Science, Marketed Products

Benefit/Risk Assessment John Affinito, MD Executive Medical Director Global Patient Safety & Evaluation, Marketed Products Clinical Perspective N. Lawrence Edwards, MD, MACP, MACR Professor and Vice Chairman Department of Medicine, University of Florida CI‐3 Consultants

Kevin J. Carroll, PhD Managing Director and Statistical Consultant KJC Statistics Ltd, UK

Andrew Whelton, MD Professor of Medicine, Adjunct Johns Hopkins University School of Medicine CI‐4 ULORIC (febuxostat) Prescribing Information Summary

ULORIC is a xanthine oxidase (XO) inhibitor indicated for the chronic management of hyperuricemia in patients with gout. ULORIC is not recommended for the treatment of asymptomatic hyperuricemia. • Approved February 2009 • 40 mg and 80 mg • 85 countries; >15 million patient‐years of exposure • Ex‐US – Other indications – Doses 10 to 120 mg CI‐5 Rationale and Benefit of Febuxostat for the Treatment of Gout

• Urate‐lowering therapy is central to long‐term management of gout – Febuxostat meets ACR recommendations to reduce and maintain serum urate • Reduces body urate pool and dissolve crystals • Decreases acute flares and reduces tophus size • Important treatment option • No dose adjustment for mild or moderate renal impairment – 40 mg for patients with severe renal impairment CI‐6 Febuxostat Clinical Development Program

Phase 2 Trials Phase 3 Trials

Study 004 APEX FACT CONFIRMS n=153 n=1072 n=760 n=2269 1 month 6 months 1 year 6 months

FOCUS EXCEL Open‐Label Open‐Label Extension Study Extension Study n=1086 n=116 3 years 5 years Adjudicated Major CV Eventsa: CI‐7 Pooled Analysis of Phase 3 Randomized Controlled Studies CONFIRMS, APEX, FACT Events, n (rate per 100 subject‐years) 95% CI Febuxostat Allopurinol n=2690 n=1277 All APTC events 10 (0.74) 4 (0.60) 0.36, 1.37 0.16, 1.53 CV death 3 (0.22) 2 (0.30) 0.05, 0.65 0.04, 1.08 Nonfatal MI 5 (0.37) 2 (0.30) 0.12, 0.87 0.04, 1.08 Nonfatal stroke 2 (0.15) 0 0.02, 0.54 0.00, 0.55 a Adjudication was done prospectively in CONFIRMS and retrospectively in APEX and FACT. CI‐8 Brief Regulatory History

• November 2008: Arthritis Advisory Committee recommended approval – Yes=12, No=0, Abstain=1 – Adequate evidence to support efficacy, safety, and quality – Cardiovascular safety risk not fully elucidated – Post‐approval study to assess CV safety • February 2009: ULORIC (febuxostat) approved – USPI included Warnings & Precautions for cardiovascular events – Post‐marketing requirement for a randomized, controlled CVOT CI‐9 Febuxostat Clinical Program

Phase 2 Trials Phase 3 Trials

Study 004 APEX FACT CONFIRMS n=153 n=1072 n=760 n=2269 1 month 6 months 1 year 6 months

FOCUS EXCEL Open‐Label Open‐Label Extension Study Extension Study n=1086 n=116 3 years 5 years

Post‐Marketing Requirement CARES (CVOT) n=6190 7 years CI‐10 What You Will Hear Today

• CARES – Primary endpoint of adjudicated MACE: febuxostat is noninferior to allopurinol • Comparable rates in non‐fatal CV events • Higher rate of CV death – Comprehensive assessment • Totality of evidence – Development program: clinical and nonclinical – Literature – Post‐marketing surveillance • Informing physicians with CARES – Label: Warnings & Precautions, Patient Information Leaflet – Dear Healthcare Provider Letter • Febuxostat is an important treatment option CL‐1

Gout: Disease Burden, Background & Treatment Landscape

Michael A. Becker, MD Professor Emeritus The University of Chicago CL‐2 Gout Is a Painful, Chronic, and Disabling Inflammatory Disorder

• Affects approximately 9 million people in the USa • Signs/Symptoms: flares, tophi, arthropathy, and urolithiasis • Saturating levels of urate in extracellular fluids may result in urate crystal formation and deposition in joints, bones, cartilage, tendons, and skinb – Hyperuricemia (sUA >6.8 mg/dL) reflects urate saturation and is necessary, but not sufficient, for the expression of gout – In the absence of urate crystal formation and tissue deposition and an inflammatory response to the crystals, the symptoms and signs of gout do not occur

a Chen‐Xu M, et al. Arthritis Rheumatol. 2019; In Press; b Doherty M, et al. Ann Rheum Dis. 2012;71:1765‐1770. Gout Flares and Monosodium Urate Crystals in Synovial Fluid CL‐3 Neutrophils on Polarized Microscopy

ABC

A. Courtesy of Michael A. Becker, MD. B. American College of Rheumatology slide collection on Rheumatic Diseases, 1997. C. Courtesy of Michael A. Becker, MD. Gout Progression May Lead to Tophi, Joint Destruction, CL‐4 and Functional Impairment

Urate deposition 3‐Dimensional Bone and joint within handsa CT imagea destructionb

a Dalbeth N, et al. Arthritis Rheumatol. 2007;56(1):29. b Image courtesy of Dr. Fernando Perez‐Ruiz. Serious Comorbidities Frequently Accompanying CL‐5 Hyperuricemia and Gout (NHANES 2007‐2008)

100% Gout 90% % No Gout 80% 70% n=5707 60% comorbidities,

50% of 40% 30% 20% Prevalence 10% 0% Hyper‐ CKD Obesity Diabetes Nephro‐ CKD MI Heart Stroke tension Stage ≥2 lithiasis Stage ≥3 failure

Zhu Y, et al. Am J Med. 2012;125(7):679‐687. CL‐6 Increased Risk of CV Death Associated With Gout

Risk of death Outcome  Lower Higher  HR (95% CI) All‐cause mortality 1.28 (1.15, 1.41)

All‐cardiovascular death 1.38 (1.15, 1.66)

Fatal coronary heart disease 1.55 (1.24, 1.93)

0.5 1 2 Relative risk (95% CI)

Choi HK, et al. Circulation. 2007;116(8):894‐900. CL‐7 Current Treatment Landscape in Gout

• Nonpharmacologic therapy: Adjunct therapy – Lifestyle adjustments and risk reduction measures (eg, weight loss, alcohol avoidance, replacement of hyperuricemia‐promoting meds) • Pharmacologic therapy – Anti‐inflammatory agents: Short‐term symptom control (not urate lowering) • NSAIDs, glucocorticoids, colchicine, IL‐1β antagonists – Urate‐Lowering Pharmacotherapy (ULT): Long‐term management • Majority of gout patients will ultimately fulfill criteria for urate‐lowering pharmacotherapy CL‐8 Urate‐Lowering Pharmacotherapy: Rationale and Support

• Urate‐Lowering Pharmacotherapy (ULT): Long‐term management – Reduce the body soluble urate pool – Dissolve deposited urate crystals – Prevent/reverse gout symptoms and progression to disability & impaired QOL • Support for urate‐lowering pharmacotherapya – 3 mechanistically distinct urate‐lowering therapeutic strategies that are disease modifyinga

a. Li‐Yu J, et al. J Rheumatol. 2001;28(3);577‐580; Perez‐Ruiz F, et al. Arthritis Rheum. 2002;47(4):356‐360; Sundy JS, et al. JAMA. 2011;306(7):711‐720; Becker MA, et al. J Rheum. 2009;36(6):1273‐1282; Shoji A, et al. Arthritis Rheum. 2004;51(3):321‐325. ACR 2012 Recommendations for CL‐9 Urate‐Lowering Pharmacotherapy

First‐line Second‐line Third‐line Xanthene oxidase Uricosuric agents Biologic uricolytic therapy inhibitors • Decrease uric acid production • Increase uric acid renal • Converts uric acid to a soluble • Available agents excretion product – Allopurinol • Available agents • Available agent – Febuxostat – Probenecid – Pegloticase – Lesinurad

Khanna D, et al. Arthritis Care Res. 2012;64(10):1431‐1467. CL‐10 ACR 2012 Gout Oral Urate‐Lowering Treatment Guidelines

• Treat to serum urate (sUA) target – sUA <6.0 mg/dL in most patients – sUA <5.0 mg/dL with persistent symptoms and/or tophi • Treatment initiation and dose titration – Initiate ULT at lowest dose of the chosen agent – Stepwise up‐titration guided by sUA determinations • Lower initiation and incremental doses in renal impairment (same goal urate levels)

Khanna D, et al. Arthritis Care Res. 2012;64(10):1431‐1467. CL‐11 Xanthine Oxidase Inhibitors

Allopurinol Febuxostat • Purine analog and non‐selective XOI • Not a purine analog but is a selective XOI • Approved in 1966 at daily doses of 100 to 800 mg • Approved in 2009 at daily doses of 40 and 80 mg • Mainstay of urate‐lowering therapy • Renal excretion <10% of dose • Excreted primarily as active metabolite oxypurinol – No need to dose reduce in mild or moderate via kidney renal impairment – 40 mg approved for patients with severe renal impairment

Allopurinol Oxypurinol

Febuxostat

De Vera MA, et al. Arthritis Care Res. 2014;66:1551‐1559. Reduction of Flares Following Treatment With Febuxostat CL‐12 vs Placebo Over 2 Years in Early Gout TMX‐67‐204

45 45

% 41.4 Placebo

40 40 Febuxostat flare, 35 35 *

gout 29.3 30 30 27.9 one 25 25 21.7 least 20 20 17.2 16.5

at ** 15 15 13.4 13.5 * 10.3 with 10 10 * 5.9 5 5

Subjects 0 0 Overall M0‐M6 M6‐M12 M12‐M18 M18‐M24 Time interval

P values for difference between treatment groups are from Fisher's exact test. *, ** indicate statistical significance at the 0.05 or 0.01 level, respectively. Dalbeth N, et al. Arthritis Rheumatol. 2017;69(12):2386‐2395. CL‐13 Lowering Serum Urate Increases Rates of Tophus Size Reduction

Allopurinol 8 Benzbromarone Combined 6 mg/dL

4 urate,

2 Serum

0 0.0 0.5 1.0 1.5 2.0 2.5 Velocity of tophus reduction, mm/month

Perez‐Ruiz F, et al. Arthritis Rheum. 2002;47(4):356‐360. CL‐14 Limited Urate‐Lowering Therapeutic Options in Gout With Many Treatment Challenges

XO Inhibitors Uricosuric Agents Pegloticase (pegylated recombinant modified Allopurinol and febuxostat Probenecid and lesinurad mammalian uricase) • Only ~50% of patients reach sUA • Both infrequently used • IV administration target at commonly used doses • Probenecid • Indicated as therapy for • Allopurinol ‒ Requires multiple daily dosing progressive gout not responsive ‒ Challenges with multiple dose ‒ Many drug‐drug interactions to conventional oral therapy titration steps to achieve target ‒ Limited efficacy in patients • Development of high titers ‒ Under‐dosed in patients with impaired renal function of anti‐pegloticase antibody with renal impairment • Lesinurad is associated with loss of urate‐ ‒ Rare, life‐threatening lowering efficacy and infusion hypersensitivity syndrome ‒ Approved only for combined reactions, including anaphylaxis therapy with an XOI • Febuxostat ‒ Not recommended in patients ‒ Cardiovascular warning with CKD3 or higher for thromboembolic events

Rashid N, et al. J Rheumatol. 2015;42(3):504‐512. Barriers to Effective Management of Gout CL‐15 Result in Low Patient and Physician Adherence

• Treatment‐initiated flares – Recurrences during early months of ULT in 40% to 50% of patients • Lag time to demonstrable clinical benefit – Improvement in symptoms not evident within the first year of treatment • Cumbersome allopurinol dose titration • Limited ULT adherence – Low flare rates early in gout often not attended to in patients with comorbidities – Lack of adherence after 1 to 2 years of successful treatment due to reduced flare rate

Becker MA, et al. N Engl J Med. 2005;353:2450‐2461. CL‐16 Summary of Gout Treatment Landscape

• Urate‐lowering therapy can relieve/reverse painful symptoms of gout and disease progression • Urate‐lowering pharmacotherapy options are limited – Renal impairment influences therapeutic choices • Xanthine oxidase inhibitors are the mainstay of urate‐lowering treatment – Allopurinol is the most prescribed; many patients receive insufficient doses • Febuxostat plays an important role in the management of gout in patients with – Renal impairment (dose adjustment not required in mild/moderate patients) – Inadequate response to or intolerance of other agents, especially allopurinol – Multiple drug‐drug interactions precluding use of other ULTs – Severe gout (sUA ≥10 mg/dL or tophi) CS‐1

CARES and Cardiovascular Safety of Febuxostat

William B. White, MD Professor of Medicine, Cardiology Center University of Connecticut School of Medicine CS‐2 Presentation Agenda

• Phase 3 CV safety data • CARES Study – Design and objectives – Primary and secondary results – Analyses to explore the CV mortality outcome • Medicare Observational Study • Post‐marketing data • Menarini’s FAST study (ongoing) CS‐3

Phase 3 Pivotal Studies CS‐4 Febuxostat Phase 3 Clinical Development Program (N=4101)

FACT APEX CONFIRMS N=760 N=1072 N=2269 1 year 6 months 6 months

Entry criteria Serum uric acid (sUA) ≥8 mg/dL, history or presence of gout

Treatment Febuxostat 80 or 120 mg Febuxostat 80, 120, 240 mg Febuxostat 40 or 80 mg Allopurinol 300 mg Allopurinol 300 mga Allopurinol 300 mga Placebo

Primary Percent of subjects with Percent of subjects with last 3 sUA levels <6 mg/dL endpoint final sUA <6 mg/dL

APEX=Allopurinol and Placebo Evaluation of FebuXostat; FACT=Febuxostat and Allopurinol Controlled Trial; CONFIRMS=CONfirmation of Febuxostat In Reducing and Maintaining Serum Urate. a Lower allopurinol dose given in renal impairment. CS‐5 Adjudicated Antiplatelet Trialists’ Collaborative (APTC) Events CONFIRMS

Patients, n (%) 95% CI Febuxostat 40 mg Febuxostat 80 mg Allopurinol n=757 n=756 n=756 All APTC events 0 3 (0.40) 3 (0.40) 0.00, 0.49 0.08, 1.16 0.08, 1.16 CV death 0 0 2 (0.26) 0.00, 0.49 0.00, 0.49 0.03, 0.95 Nonfatal MI 0 1 (0.13) 1 (0.13) 0.00, 0.49 <0.01, 0.74 <0.01, 0.74 Nonfatal stroke 0 2 (0.26) 0 0.00, 0.49 0.03, 0.95 0.00, 0.49

Allopurinol dose (300 or 200 mg) based on renal function. n=145 for 200 mg (estimated CrCL 30 to 59 mL/min). Adjudicated Major CV Eventsa: CS‐6 Pooled Analysis of Phase 3 Randomized Controlled Studies CONFIRMS, APEX, FACT Events, n (rate per 100 patient‐years) 95% CI Febuxostat Allopurinol n=2690 n=1277 All APTC events 10 (0.74) 4 (0.60) 0.36, 1.37 0.16, 1.53 CV death 3 (0.22) 2 (0.30) 0.05, 0.65 0.04, 1.08 Nonfatal MI 5 (0.37) 2 (0.30) 0.12, 0.87 0.04, 1.08 Nonfatal stroke 2 (0.15) 0 0.02, 0.54 0.00, 0.55 a Adjudication was done prospectively in CONFIRMS and retrospectively in APEX and FACT. CS‐7 Cardiovascular Safety Observations in Phase 3 Pivotal Studies

• Small differences in CV events based on 14 APTC events • Prospectively adjudicated CV events from largest pivotal trial showed no imbalances (CONFIRMS, N=2269) – Results were not conclusive • CARES CV outcome trial conducted per FDA post‐marketing requirement CS‐8

CARES Study Cardiovascular Safety of Febuxostat and Allopurinol in Subjects With Gout and Cardiovascular Morbidities CS‐9 Study Design CARES

• Multicenter, double‐blind, randomized study of febuxostat vs allopurinol (once daily) in patients with gout and CV disease in the US, Canada, and Mexico

Febuxostat QD (40‐80 mg; n=3101)

Treatment period (up to 7 years) Allopurinol QDa (200‐600 mg; n=3097)

Days –7 Day 1 Month 1Month 3 Month 6 Month 12 End of to 1 study Blinded dose titration Visits every 6 months Screening Baseline/ visit randomization a At randomization, patients were stratified according to renal function. After randomization, dose titration of allopurinol was made on the basis of renal function. Febuxostat did not require dose adjustment by renal function. CS‐10 Drug Titration Designed to Achieve Serum Urate Level <6 mg/dL CARES

FEBUXOSTAT ALLOPURINOL

600 mg 80 mg 500 mg 400 mg 400 mg

40 mg 300 mg 300 mg 200 mg

Initial Dose escalate Initial Dose escalate Initial Dose escalate dose if sUA ≥6.0 mg/dL dose if sUA ≥6.0 mg/dL dose if sUA ≥6.0 mg/dL after 2 weeks Week 2‐12 Week 2‐12 Normal or mildly impaired Moderately impaired renal function renal function

Normal or Mildly impaired renal function: CrCL ≥60 mL/min. Moderately impaired renal function: CrCL 30‐59 mL/min. CS‐11 Key Entry Criteria CARES

Inclusion Exclusion • Male ≥50 yr and females ≥55 yr and ≥2 years • History of MI or stroke within 60 days post‐menopausal prior to the screening visit • History or presence of gouta • Secondary hyperuricemia – sUA level ≥7.0 mg/dL, or ≥6.0 mg/dL with • Received urate‐lowering therapy or inadequately controlled gout excluded medication <7 days prior to • History of major CV or cerebrovascular disease Study Day 1/Randomization visit including at least ONE of the following • Severe kidney disease – Myocardial infarction (estimated CrCL <30 mL/min) – Hospitalized unstable angina – Cardiac or cerebral revascularization procedure – Stroke – Hospitalized transient ischemic attack – Peripheral vascular diseaseb

– History of diabetes mellitus with a Defined as having one or more of the American Rheumatism Association criteria. micro‐ or macro‐vascular diseasec b Defined as ankle brachial index ≤0.6, revascularization and/or a well‐documented history of claudication. c Including retinopathy, neuropathy, nephropathy, small vessel vascular diseases. CS‐12 Primary and Secondary Endpoints CARES

Primary Secondary • Time from randomization to first • Time from randomization to the occurrence of the following Major occurrence of each individual Adverse Cardiovascular Events event in the MACE composite (MACE) composite endpoint • Time from randomization to first – Cardiovascular (CV) death occurrence of the Antiplatelet – Non‐fatal myocardial infarction Trialists’ Collaborative (APTC) – Non‐fatal stroke composite of – Unstable angina with urgent – CV death coronary revascularization – Non‐fatal myocardial infarction – Non‐fatal stroke CS‐13 Endpoint and Safety Monitoring Committees CARES

CARES Cardiovascular Endpoints (CEC) Committee • All suspected serious cardiovascular events and deaths were assessed by CV expert clinicians blinded to treatment assignment CARES Data Monitoring Committee (DMC) • Independent DMC unblinded to treatment assignment reviewed study results at 6‐month intervals • Performed 3 planned interim analyses (accrual of approximately 25%, 50%, and 75% of events) • Data tables were presented by an independent statistician CS‐14 Statistical Analyses CARES

• Primary endpoint – Cox proportional hazards (PH) model of time to primary endpoint stratified by baseline renal function – One‐sided, adjusted 97.5% CI calculated for hazard ratio (HR) • Interim analyses for primary endpoint after 25%, 50%, and 75% of events – Lan‐DeMets‐O’Brien‐Fleming‐alpha spending function used to control overall one‐sided significance level of 0.025 • Non‐inferiority concluded if upper bound of repeated confidence interval for HR (febuxostat to allopurinol) <1.3 • Secondary and major exploratory endpoints – Cox PH model of time to endpoint stratified by baseline renal function with two‐sided 95% CI for HR CS‐15

CARES CV Safety Results CS‐16 Baseline Demographics and Characteristics CARES

Febuxostat Allopurinol n=3098 n=3092 Mean age, years (SD) 64.6 (8.6) 65.0 (8.5) Gender male, n (%) 2604 (84.1) 2592 (83.8) Race, n (%) White 2160 (69.7) 2140 (69.2) Black 552 (17.8) 593 (19.2) Mean BMI, kg/m2 (SD) 33.6 (7.0) 33.4 (6.9) ≥30, n (%) 2053 (66.3) 2024 (65.5) Renal function, n (%)a Moderately impaired (30‐59 mL/min) 1636 (52.8) 1631 (52.7) Mildly impaired (60‐89 mL/min) 1217 (39.3) 1231 (39.8) Normal (≥90 mL/min) 239 (7.7) 228 (7.4) Use of low‐dose aspirin (<325 mg per day), n (%) 1496 (48.3) 1481 (47.9)

a Creatinine clearance (CrCL): mL/min (Cockroft‐Gault). CS‐17 Baseline Serum Urate Levels and Gout History CARES

Febuxostat Allopurinol n=3098 n=3092 Mean baseline sUA, mg/dL (SD) 8.74 (1.7) 8.69 (1.7)

sUA ≥10 mg/dL, n (%) 654 (21.1) 631 (20.4)

Mean years from onset of gout (SD) 11.8 (11.4) 11.9 (11.2)

Baseline tophus, n (%) 668 (21.6) 650 (21.0) CS‐18 Cardiovascular Disease at Baseline CARES

Patients, n (%)a Febuxostat Allopurinol n=3098 n=3092 Myocardial infarction 1197 (38.6) 1231 (39.8) Stroke 460 (14.8) 410 (13.3) Hospitalized unstable angina 855 (27.6) 869 (28.1) Hospitalized transient ischemic attack 362 (11.7) 291 (9.4) Cardiac revascularization 1129 (36.4) 1182 (38.2) Cerebrovascular revascularization 69 (2.2) 54 (1.7) Peripheral vascular disease 412 (13.3) 375 (12.1) Diabetes mellitus with small‐vessel disease 1193 (38.5) 1213 (39.2)

a Patients may have had ≥1 CV disease. Baseline Treatments for Gout and Cardiovascular Disease CS‐19 According to Treatment Group CARES Patients, n (%) Febuxostat Allopurinol n=3098 n=3092 Prior urate‐lowering therapies None 1045 (33.7) 1044 (33.8) Febuxostat 134 (4.3) 130 (4.2) Allopurinol 1738 (56.1) 1742 (56.3) Probenecid 37 (1.2) 36 (1.2) Other 5 (0.2) 6 (0.2) Cardiovascular therapies Aspirin 1839 (59.4) 1877 (60.7) Clopidogrel 558 (18.0) 596 (19.3) Beta‐blocking agents 1796 (58.0) 1834 (59.3) Lipid‐lowering therapies 2293 (74.0) 2281 (73.8) Renin‐angiotensin blocking agents 2144 (69.2) 2187 (70.7) Diuretics 1302 (42.0) 1242 (40.2) CS‐20 Final Dose Used in Trial by Treatment Group CARES

Febuxostat Allopurinola n=3098 n=3092 Final dose Patients, n (%) Final dose Patients, n (%) 40 mg 1890 (61.0) 200 mg 674 (21.8) 80 mg 1208 (39.0) 300 mg 1380 (44.6) 400 mg 778 (25.2) 500 mg 132 (4.3) 600 mg 128 (4.1)

a Maximal dose in patients with CrCL <60 mL/min was 400 mg. CS‐21 Patient Disposition CARES

Patients, n (%) Febuxostat Allopurinol Randomized, n 3101 3097 Randomized but did not take study drug, na 35 Completed study drugb 1324 (42.7) 1365 (44.1) Discontinued study drug 1777 (57.3) 1732 (55.9) Voluntary withdrawal 699 (22.5) 670 (21.6) Adverse events 453 (14.6) 446 (14.4) Other (eg, AEs; withdrawn by PI, site closure) 300 (9.7) 321 (10.4) Completed study visitsb 1704 (55.0) 1706 (55.1) Discontinued study visits 1397 (45.0) 1391 (44.9) Voluntary withdrawal 595 (19.2) 587 (19.0) Other (eg, AEs; site closure) 333 (10.7) 363 (11.7) Lost to follow‐up 226 (7.3) 223 (7.2) a Excluded from the modified intention‐to‐treat (mITT) population. b Baseline characteristics of patients who discontinued were similar to those who completed. Baseline Characteristics for Those Who Completed and CS‐22 Those Who Failed to Complete Study Visits Were Comparable CARES

All Randomized Patients, n (%) Febuxostat Allopurinol Completed Failed to complete Completed Failed to complete study visits study visits study visits study visits Characteristic n=1950 n=1151 n=1924 n=1173 Median age, years (IQR) 64.0 (58, 70) 64.0 (58, 71) 65.0 (58, 71) 65.0 (58, 71) Age ≥65 yr , n (%) 956 (49.0) 560 (48.7) 997 (51.8) 592 (50.5) Male, n (%) 1652 (84.7) 954 (82.9) 1609 (83.6) 986 (84.1) Mean duration of gout, yr (SD) 12.1 (11.6) 11.2 (11.2) 11.9 (11.0) 11.8 (11.5) Presence of tophi, n (%) 453 (23.2) 215 (18.7) 427 (22.2) 224 (19.1) Mean baseline serum uric acid (SD) 8.7 (1.6) 8.8 (1.8) 8.6 (1.6) 8.8 (1.7) Median body weight, kg (IQR) 97.2 (84, 113) 98.8 (84, 114) 96.8 (84, 112) 97.7 (85, 115) Mean BMI (SD) 33.5 (6.9) 33.6 (7.1) 33.4 (6.8) 33.5 (7.1)

White WB, et al. N Engl J Med. 2018;378:1200‐1210. Baseline Characteristics for Those Who Completed and CS‐23 Those Who Failed to Complete Study Visits Were Comparable CARES

All Randomized Patients, n (%) Febuxostat Allopurinol Completed Failed to complete Completed Failed to complete study visitsa study visits study visitsa study visits Characteristic n=1950 n=1151 n=1924 n=1173 Cardiovascular risk factors and history Diabetes mellitus with small vessel disease 750 (38.5) 444 (38.6) 751 (39.0) 464 (39.6) Hypertension 1803 (92.5) 1063 (92.4) 1755 (91.2) 1101 (93.9) Hyperlipidemia 1686 (86.5) 993 (86.3) 1683 (87.5) 1023 (87.2) Myocardial infarction 758 (38.9) 441 (38.3) 747 (38.8) 485 (41.3) Hospitalization for unstable angina 527 (27.0) 328 (28.5) 529 (27.5) 341 (29.1) Moderate renal impairment 998 (51.2) 639 (55.5) 1007 (52.3) 628 (53.5) (eCrCL <60 mL/min and ≥30 mL/min), n (%) Stroke 280 (14.4) 180 (15.6) 241 (12.5) 171 (14.6) Congestive heart failure 367 (18.8) 256 (22.2) 348 (18.1) 285 (24.3) a Patients who completed study visits or had a primary MACE endpoint observed while on study. White WB, et al. N Engl J Med. 2018;378:1200‐1210. CS‐24 Primary Endpoint: Non‐inferiority for MACE CARES

Patients, n (%) Febuxostat Allopurinol Hazard ratioa n=3098 n=3092 (upper bound CI) 75% interim analysis 244 (7.9) 244 (7.9) 0.99 (1.23) (primary analysis)b Final analysis 335 (10.8) 321 (10.4) 1.03 (1.23)c

Composite MACE endpoint: CV death, nonfatal MI, nonfatal stroke, unstable angina with urgent coronary revascularization

a Febuxostat to allopurinol. b n=3039 and 3034 respectively. c P=0.002 for non‐inferiority. White WB, et al. N Engl J Med. 2018;378(13):1200‐1210. CS‐25 Time to First Occurrence of the Primary Endpoint CARES Final Analysis

25 Febuxostat (95% CI) Allopurinol (95% CI) HR=1.03 (0.87, 1.23) 20 % 15

10 Cumulative,

5

0 0 6 12 18 24 30 36 42 48 54 60 66 72 Time, months Patients at risk, n Febuxostat 3098 2784 2493 2111 1854 1589 1369 1165 955 778 573 441 264 Allopurinol 3092 2764 2465 2080 1815 1560 1361 1132 933 767 589 437 258 CS‐26 Secondary Endpoints: Components of MACE composite CARES

Febuxostat Allopurinol Absolute risk n=3098 n=3092 difference Hazard ratioa (95% CI) P valueb

Composite primary endpoint 335 (10.8) 321 (10.4) 0.4% 1.03 (0.87, 1.23)c 0.66

CV death 134 (4.3) 100 (3.2) 1.1% 1.34 (1.03, 1.73) 0.03

Nonfatal MI 111 (3.6) 118 (3.8) ‐0.2% 0.93 (0.72, 1.21) 0.61

Nonfatal stroke 71 (2.3) 70 (2.3) 0 1.01 (0.73, 1.41) 0.94

Unstable angina with urgent 49 (1.6) 56 (1.8) ‐0.2% 0.86 (0.59, 1.26) 0.45 coronary revascularization

a Febuxostat to allopurinol. b P values are for the test of superiority of febuxostat to allopurinol and were calculated with the use of a Cox regression analysis. c Adjusted for interim analyses. White WB, et al. N Engl J Med. 2018;378:1200‐1210. CS‐27 Time to CV Death CARES

25 Febuxostat (95% CI) Allopurinol (95% CI) HR=1.34 (1.03, 1.73) 20 % 15

10 Cumulative,

5

0 0 6 12 18 24 30 36 42 48 54 60 66 72 Patients at risk, n Time, months Febuxostat 3098 2823 2550 2174 1922 1659 1440 1243 1033 838 627 482 288 Allopurinol 3092 2807 2530 2152 1898 1637 1433 1204 1008 838 646 489 287 Sensitivity Analysis: CV Events on Drug or Within 30 Days After CS‐28 Drug Discontinuation CARES

Febuxostat Allopurinol Absolute risk n=3098 n=3092 difference Hazard ratioa (95% CI) P value

Composite primary endpoint 242 (7.8) 238 (7.7) 0.1% 1.00 (0.82, 1.22)b 0.99

CV death 62 (2.0) 41 (1.3) 0.7% 1.49 (1.01, 2.22) 0.047

Nonfatal MI 93 (3.0) 106 (3.4) ‐0.4% 0.87 (0.66, 1.15) 0.32

Nonfatal stroke 59 (1.9) 62 (2.0) ‐0.1% 0.94 (0.66, 1.34) 0.72

Unstable angina with urgent 45 (1.5) 44 (1.4) 0 1.00 (0.66, 1.52) 0.99 coronary revascularization

a Febuxostat to allopurinol. b Adjusted for interim analyses. CS‐29 All‐Cause Mortality (mITT Analysis) CARES

Patients, n (%) Febuxostat Allopurinol n=3098 n=3092 Hazard ratioa (95% CI) All‐cause mortality 243 (7.8) 199 (6.4) 1.22 (1.01, 1.47) Vital status collection blinded to 1159 1193 treatment assignmentb Additional deaths 89 110 All‐cause mortality with 332 (10.7) 309 (10.0) 1.09 (0.94, 1.28) additional vital status data

a Febuxostat to allopurinol. b Conducted towards the latter part of the study by OmniTrace in collaboration with study sites. CS‐30 Other Analyses to Explore Cardiovascular Mortality Outcome CARES

• Adjudicated causes of death • Subgroup analyses • Vital signs • Electrolytes • Mechanistic studies • Association with serum urate and gout flares CS‐31 Adjudicated Causes of Death CARES

Patients, n (%) Febuxostat Allopurinol n=3098 n=3092 Cardiovascular deaths 134 (4.3) 100 (3.2) Sudden cardiac death 83 (2.7) 56 (1.8) Heart failure 20 (0.6) 13 (0.4) Stroke 8 (0.3) 11 (0.4) Myocardial infarction 11 (0.4) 6 (0.2) Cardiac arrhythmia 7 (0.2) 9 (0.3) Valvular heart disease 3 (<0.1) 2 (<0.1) Heart and respiratory failure 1 (<0.1) 1 (<0.1) Cardiovascular hemorrhage 0 1 (<0.1) Peripheral arterial disease 0 1 (<0.1) Other cardiovascular deaths 1 (<0.1) 0 Non‐cardiovascular deaths 109 (3.5) 99 (3.2) CS‐32 Risk Ratios (95% CI) for CV Death by Subgroups CARES

Interaction Baseline variables  Lower risk on febuxostat Lower risk on allopurinol  RR (95% CI) P value Overall mITT 1.34 (1.04, 1.72) Age, yr <65 1.08 (0.70, 1.68) 0.205 ≥65 1.52 (1.12, 2.08) Gender Female 1.32 (0.65, 2.70) 0.971 Male 1.34 (1.02, 1.76) BMI, kg/m2 <30 1.37 (0.91, 2.08) 0.863 ≥30 1.31 (0.95, 1.81) Race White 1.38 (1.03, 1.83) 0.636 Non‐white 1.19 (0.69, 2.06) Years since gout diagnosis <5 1.33 (0.91, 1.93) 0.508 5‐10 0.94 (0.47, 1.86) >10 1.49 (1.00, 2.24) Baseline sUA, mg/dL <9.0 1.75 (1.16, 2.64) 0.225 9.0‐10.0 1.04 (0.62, 1.76) ≥10.0 1.14 (0.75, 1.72) NSAID use Yes 2.17 (1.28, 3.68) 0.032 No 1.13 (0.84, 1.52) Low‐dose aspirin use Yes 0.85 (0.58, 1.24) 0.001 No 1.99 (1.39, 2.85) Colchicine used during study Yes 2.17 (1.21, 3.89) 0.062 No 1.18 (0.89, 1.57)

0.25 0.5 1 2 4 Risk ratio (95% CI) CS‐33 Risk Ratios (95% CI) for CV Death by Subgroups CARES

Interaction Baseline variables  Lower risk on febuxostat Lower risk on allopurinol  RR (95% CI) P value Overall mITT 1.34 (1.04, 1.72) History of nonfatal MI Yes 1.34 (0.96, 1.85) 0.955 No 1.37 (0.92, 2.05) History of nonfatal stroke Yes 2.27 (1.14, 4.50) 0.090 No 1.21 (0.92, 1.60) History of cardiac revascularization Yes 1.46 (1.03, 2.07) 0.528 No 1.25 (0.86, 1.80) History of congestive heart failure Yes 1.13 (0.79, 1.60) 0.201 No 1.60 (1.11, 2.30) History of hyperlipidemia Yes 1.24 (0.95, 1.63) 0.169 No 2.14 (1.03, 4.43) History of diabetes Yes 1.26 (0.90, 1.75) 0.585 No 1.46 (0.98, 2.16) History of hypertension Yes 1.34 (1.03, 1.74) 0.888 No 1.24 (0.38, 3.99) Renal function Moderate 1.19 (0.89, 1.59) 0.379 Mild 1.81 (1.04, 3.16) Normal 1.91 (0.48, 7.54) 0.25 0.5 1 2 4 8 Risk ratio (95% CI) Effects of Febuxostat and Allopurinol on CS‐34 Blood Pressure and Heart Rate CARES

Systolic BP Diastolic BP Heart rate

140 Febuxostat 90 80

138 Allopurinol 88 78

136 86 76

134 84 74 mmHg

mmHg bpm

132 82 72 rate, 130 80 70 pressure,

pressure,

heart 128 78 68 blood

blood

126 76 Mean 66 Mean

124 Mean 74 64

122 72 62

120 70 60 0 6 12 18 24 30 36 42 48 54 60 66 72 0 6 12 18 24 30 36 42 48 54 60 66 72 0 6 12 18 24 30 36 42 48 54 60 66 72 Time, months Time, months Time, months Effects of Febuxostat and Allopurinol on CS‐35 Serum Potassium and Calcium CARES

Serum calcium Serum potassium 10 5 mEq/L

mg/dL

9 4 calcium, potassium,

serum serum

Mean Febuxostat Mean Febuxostat Allopurinol Allopurinol 8 3 0 6 12 18 24 30 36 42 48 54 60 66 72 0 6 12 18 24 30 36 42 48 54 60 66 72 Time, months Time, months Mechanistic Studies to Evaluate the Imbalance in CS‐36 Sudden CV Death on Febuxostat Versus Allopurinol

• No effects of febuxostat on – Cardiac purkinje fibers and hERG assay – Sodium or calcium channels – Platelet function, INR, and PTT – Cardiac conduction (thorough QTc study) Thorough QT Study of Febuxostat at Maximal Therapeutic and CS‐37 Supra‐Therapeutic Doses in Healthy Volunteers

Comparison of maximum QTFC intervals

Mean of maximum QTFC, msec Difference from placebo, msec Febuxostat Febuxostat Moxifloxacin Febuxostat Febuxostat Moxifloxacin Placebo 80 mg 300 mg 400 mg 80 mg 300 mg 400 mg Baseline 406.0 407.1 405.7 404.3 1.1 ‐0.3 ‐1.7

Day 1 396.5 398.6 398.5 406.8a 2.1 2.0 10.3

Day 4 402.9 403.7 405.1 415.9a 0.8 2.2 13.0

Randomized, double‐blind, crossover study, N=41. FC=Fridericia corrected QT interval. a Indicates statistically significant difference from placebo at the 0.001 level. Cardiovascular Mortality Not Related to Serum Urate Levels CS‐38 (Patients With CV Death Compared to All Other Patients) CARES

12 No CV mortality Febuxostat (n=2964) Allopurinol (n=2992) 11 CV mortality Febuxostat (n=134) Allopurinol (n=100) 10 9 mg/dL 8 sUA,

7 6 Mean 5 4 3 0123456 1218243036424854606672 Time, months

Note: There were too few fatal events after 48 months to calculate precise urate levels. Cardiovascular Mortality Not Related to Gout Flare Rates CS‐39 (Patients With CV Death Compared to All Other Patients) CARES 1.8 No CV mortality Febuxostat (n=2964) Allopurinol (n=2992) 1.6 CV mortality Febuxostat (n=134) Allopurinol (n=100) 1.4 year ‐ 1.2 patient

1.0 per 0.8 flares

0.6

Gout 0.4

0.2

0.0

‐0.2 0 to <6 6 to <12 12 to <18 18 to <24 24 to <30 30 to <36 36 to <42 42 to <48 48 to <54 54 to <60 60 to <66 66 to <72 Time interval, months

Data on file. CS‐40 Conclusions on the Cardiovascular Safety Findings CARES

• Strength: Large, randomized, double‐blind, controlled study • Febuxostat non‐inferior to allopurinol for primary composite MACE • Similar rates of nonfatal MI, nonfatal stroke, and unstable angina with urgent revasc • Rate of all‐cause mortality higher on febuxostat: driven by increases in CV death • Limitation: Large drop‐out rates in both treatment groups – Sensitivity analyses consistent to mITT – Baseline characteristics of those who dropped out similar to those who completed the trial • Subgroup analyses showed no relevant heterogeneity for treatment effect • Febuxostat and allopurinol exhibited similar effects on clinical laboratory test results, vital signs, and gout flare rates • No mechanism for febuxostat increasing death due to CV causes has been identified CS‐41 Additional Evidence

• Other published studies – Medicare Observational Study • Post‐marketing data • Ongoing Febuxostat vs Allopurinol Streamlined Trial (FAST) CS‐42 Medicare Observational Study Design

• Observational study based on Medicare claims data (>99,000 subjects) • Population: age ≥65 with gout – Subjects selected based on ICD9 codes for gout diagnosis – Propensity score matched initiators of febuxostat or allopurinol based on National Drug Code • Events identified through validated claims‐based algorithm – Positive predictive value >80% • Primary and secondary outcomes – Hospitalization for MI or stroke – All‐cause mortality captured; causes of death not possible to determine

Zhang M, et al. Circulation. 2018;138:1116‐1126. Medicare Study—Risk of Cardiovascular Events in Febuxostat vs CS‐43 Allopurinol Initiators: 1:3 PS‐Matched Analysis

Incidence ratea (95% CI) Febuxostat Allopurinol Lower risk with Outcome n=24,936 n=74,808  Febuxostat Allopurinol  HR (95% CI) As‐treated analysis Primary outcome MI or stroke 3.43 (3.22, 3.66) 3.36 (3.25, 3.49) 1.01 (0.94, 1.08) Secondary outcomes MI 2.17 (2.00, 2.35) 2.08 (1.99, 2.18) 1.03 (0.94, 1.13) Stroke 1.35 (1.22, 1.49) 1.36 (1.29, 1.44) 0.98 (0.87, 1.10) Coronary revascularization 2.64 (2.46, 2.84) 2.75 (2.65, 2.86) 0.95 (0.87, 1.03) All‐cause mortality 4.11 (3.88, 4.36) 4.27 (4.14, 4.40) 0.95 (0.89, 1.02)

0.71.0 1.4 Hazard ratio (95% CI)

a Incidence rate per 100 person‐yr. Zhang M, et al. Circulation. 2018;138:1116‐1126. Medicare Study—Subgroup Analysis by Baseline Cardiovascular CS‐44 Disease: 1:3 PS‐Matched As‐Treated Analysis

Incidence ratea (95% CI) Lower risk with Febuxostat Allopurinol Febuxostat Allopurinol  HR (95% CI) Without baseline CVD n=21,821 n=65,463 Primary outcome MI or stroke 2.96 (2.75, 3.18) 2.96 (2.85, 3.08) 0.99 (0.91, 1.08) Secondary outcomes MI 1.85 (1.69, 2.03) 1.88 (1.79, 1.98) 0.98 (0.88, 1.08) Stroke 1.17 (1.05, 1.32) 1.16 (1.09, 1.23) 1.00 (0.88, 1.15) Coronary revascularization 2.36 (2.17, 2.56) 2.53 (2.43, 2.64) 0.92 (0.84, 1.01) All‐cause mortality 3.79 (3.55, 4.04) 3.89 (3.76, 4.03) 0.97 (0.90, 1.04) With baseline CVD n=3067 n=9201 Primary outcome MI or stroke 7.80 (6.80, 8.93) 7.87 (7.31, 8.48) 0.97 (0.83, 1.13) Secondary outcomes MI 5.18 (4.38, 6.11) 5.16 (4.71, 5.65) 0.98 (0.81, 1.19) Stroke 2.90 (2.32, 3.61) 2.77 (2.45, 3.14) 1.03 (0.80, 1.32) Coronary revascularization 5.55 (4.72, 6.52) 4.93 (4.49, 5.42) 1.10 (0.91, 1.33) All‐cause mortality 7.05 (6.13, 8.12) 8.19 (7.62, 8.79) 0.85 (0.72, 0.99) 0.71.0 1.4 Hazard ratio (95% CI) a Incidence rate per 100 person‐yr. Zhang M, et al. Circulation. 2018;138:1116‐1126. CS‐45 Conclusions: Medicare Observational Study

• Strengths – Large dataset that is well characterized – Represents an elderly gout population in clinical practices in the US – Propensity score matching performed by experienced analytical team • Limitations – Observational study with imbalance in exposure – Nonrandomized population who are older but had similar CV disease risk – No data on CV death • Conclusion – No evidence of increased CV risk with febuxostat compared with allopurinol in subjects with or without underlying CV disease

Zhang M, et al. Circulation. 2018;138:1116‐1126. No Signal for CV Events or CV Death With Febuxostat Observed CS‐46 Through Pharmacovigilance Activities

• Estimated patient exposure – Global: 15 million patient‐years – United States: 1.4 million patient‐years • Postmarketing surveillance – Ongoing monitoring of Individual Case Safety Reports (ICSRs) received – Weekly literature reviews – Aggregate review of ICSRs in the global safety database • Takeda data mining parameters (reviewed quarterly): – Events reported ≥ 3 AND EB05a ≥ 2

a Lower of the 90% confidence limit of the Empirical Bayes Geometric Mean. CS‐47 Febuxostat vs Allopurinol Streamlined Trial (FAST) Sponsored by Menarini International

• Prospective, randomized, open, blinded endpoint study (PROBE) – ~6000 gout patients with at least one major CVD risk factora – Treat to sUA target (<357 µmol/L) with allopurinol • Randomized 1:1 to febuxostat (80 or 120 mg) or allopurinol (optimized dose) • Primary endpoint: APTC composite events (n=456 events) • Average follow‐up ~3 years (expected to complete mid‐2020) Allopurinol (at optimized dose) Screening/eligibility sUA YES Randomize (prior allopurinol) <357 µmol/L (N=5706) Febuxostat NO (80 mg daily, increase to 120 mg if sUA above target) Optimize allopurinol dose a Major CVD risk factors: Age ≥70 years (male) or ≥75 years (female); smoking, FHx, dyslipidemia, HTN, DM, impaired glucose tolerance, obesity; previous CVD (MI, CVA or TIA); peripheral vascular disease; CKD, microalbuminuria or proteinuria; inflammatory arthritis (RA, PsA, and AS) or chronic NSAID use; COPD. MacDonald TM, et al. BMJ Open. 2014;4(7):e005354. CS‐48 Totality of Evidence Regarding CV Safety of Febuxostat

• Imbalance in CV death in CARES – Not previously observed in CVOTs of noncardiac drugs when nonfatal CV events balanced – Main between‐group difference was in sudden cardiac death – No differences in adjudicated arrhythmias (fatal or nonfatal) • Clinical and biochemical variables do not explain difference in CV death – Cardiac channel and prothrombotic mechanisms not observed • Without placebo group, impact of xanthine oxidase inhibitors on risk of CV mortality cannot be defined • Large Medicare observational study (>99,000 subjects) found no differences in morbidity or mortality between febuxostat and allopurinol • No signal for CV events or CV death from post‐marketing surveillance • Additional data forthcoming from FAST CE‐1

Efficacy of Febuxostat

Lhanoo Gunawardhana, MD, PhD Senior Medical Director Clinical Science, Marketed Products CE‐2 Febuxostat Clinical Program

Phase 2 Trials Phase 3 Trials

Study 004 APEX FACT CONFIRMS n=153 n=1072 n=760 n=2269 1 month 6 months 1 year 6 months

FOCUS EXCEL Open‐Label Open‐Label Extension Study Extension Study n=1086 n=116 3 years 5 years

Post‐Marketing Requirement CARES (CVOT) n=6198 7 years Key Differences in Design of Phase 3 Development Studies CE‐3 and CARES Study Relevant to Efficacy Assessment

• Phase 3 studies for development program – Optimized for efficacy assessment • Primary endpoint based on achieving sUA <6.0 mg/dL • sUA entry criteria ≥8.0 mg/dL – Febuxostat compared with most commonly used doses of allopurinol (100 to 300 mg depending on renal function) • CARES Study – Designed as safety study • Efficacy assessed as exploratory endpoints • sUA entry criteria ≥7.0 mg/dL, or ≥6.0 mg/dL in pts with frequent flares or tophi – Dose of febuxostat and allopurinol titrated to target sUA <6.0 mg/dL CE‐4 Demographics CONFIRMS, APEX, FACT

CONFIRMS APEX FACT n=2269 n=1072 n=760 Male, n (%) 2141 (94) 1005 (94) 729 (96) Race, n (%) Caucasian 1863 (82) 835 (78) 587 (77) Black 228 (10) 120 (11) 62 (8) Asian 88 (4) 26 (2) 25 (3) Mean age, yr 52.8 51.6 51.8 Mean BMI, kg/m2 32.8 32.7 32.5 Presence of tophi, n (%) 478 (21) 299 (28) 186 (24) Mean yr with gout 11.6 10.9 11.9 Mean baseline sUA, mg/dL 9.6 9.9 9.8 CE‐5 Medical History CONFIRMS, APEX, FACT

Patients, n (%) CONFIRMS APEX FACT Medical condition n=2269 n=1072 n=760 Renal functiona eCrCL, mL/minb 30‐59 (moderately impaired) 402 (18) 154 (14) 94 (12) 60‐89 (mildly impaired) 1081 (48) 377 (35) 295 (39) ≥90 (normal) 786 (35) 541 (51) 371 (49) Hypertension 1199 (53) 502 (47) 331 (44) Hyperlipidemia 942 (42) 349 (33) 255 (34) Diabetes 312 (14) 90 (8) 53 (7) Atherosclerotic disease 261 (12) 145 (14) 75 (10) Use of low‐dose aspirin <325 mg/day 405 (18) 183 (17) 128 (17) a Estimated creatinine clearance (eCrCL): mL/min (Cockroft‐Gault). b Kidney Disease Outcome Quality Initiative; Chronic Kidney Disease. Am J Kidney Dis. 2002;39:S46‐S75. sUA <6 mg/dL at Final Visit in CE‐6 Overall Population CONFIRMS, APEX, FACT

100 Febuxostat 40 mg Febuxostat 80 mg Allopurinol

80 72* 74* 67*# % 60 45‡ 42 39 Subjects, 40 36

20

n= 757 756 755 253 263 249 242 0 CONFIRMS APEX FACT

* P<0.001 vs allopurinol. # P<0.001 vs febuxostat 40 mg. ‡ Non‐inferior to allopurinol. sUA <6 mg/dL at Final Visit in Patients With CE‐7 Mild/Moderate Renal Impairment CONFIRMS, APEX, FACT

100 Febuxostat 40 mg Febuxostat 80 mg Allopurinol

80 78 * 77* 72*, ** %

60 50* 45 42 43 40 Subjects,

300/ 300/ 300 mg 20 200 mg 100 mg

n= 479 503 501 126 135 128 121 0 CONFIRMS APEX FACT

CrCL >30 to <90 mL/min. * P<0.05 vs allopurinol. ** P<0.05 vs febuxostat 40 mg. sUA <6 mg/dL at Final Visit in Patients With CE‐8 Baseline sUA ≥10 mg/dL CONFIRMS, APEX, FACT

100 Febuxostat 40 mg Febuxostat 80 mg Allopurinol

80 67* 60* %

60 49*#

40 Subjects, 31 27 21 21 20

n= 249 254 230 103 89 104 103 0 CONFIRMS APEX FACT

* P<0.05 vs allopurinol. # P<0.05 vs febuxostat 40 mg. sUA <6 mg/dL at Final Visit in Patients With CE‐9 Tophi at Baseline CONFIRMS, APEX, FACT

100 Febuxostat 40 mg Febuxostat 80 mg Allopurinol

80 67*

#

% 57* * 60 56

40 Subjects, 35 35 32 30

20

n= 166 163 148 43 63 52 46 0 CONFIRMS APEX FACT

* P<0.05 vs allopurinol. # P<0.05 vs febuxostat 40 mg. CE‐10 Febuxostat Reduces Gout Flares Over Time Long‐Term Open‐Label Extension Studies FOCUS, EXCEL

50 • 80% maintained sUA <6 mg/dL on febuxostat – ~80% remained on 80 mg

% 40 • Tophi resolved in ~50% of subjects after 2 years flare, 30 with FOCUS 20 EXCEL Subjects

10

0 0‐66‐12 12‐18 18‐24 24‐30 30‐36 36‐42 42‐48 48‐54 54‐60 >60 Time interval, mo Clinical Efficacy of Febuxostat vs Allopurinol in CE‐11 Real‐World Setting (Proportion of Subjects That Achieved Target sUA)

Propensity score‐matched Propensity score‐matched treatment‐naive new users (n=1746) new and current users (n=3864) 70 70 Febuxostat Febuxostat 58.5 60 56.9 Allopurinol 60 Allopurinol

47.1* 50 44.8* 50 goal goal 40 35.5 40 36.2 sUA sUA 30 30 at at

% % 21.8* 19.2* 20 20

10 10

0 0 <6.0 mg/dL <5.0 mg/dL <6.0 mg/dL <5.0 mg/dL Post‐index sUA Post‐index sUA

* P<0.001 febuxostat vs allopurinol. The most common doses were 40 mg/day for febuxostat (83%), and 300 mg/day or lower dose for allopurinol (97%). Singh JA, et al. Arthritis Research & Therapy. 2015;17:120. CE‐12

Efficacy in CARES Proportion of Patients With Serum Urate Levels <6.0 mg/dL CE‐13 by Visit (FAS) CARES

Overall population Moderate renal impairment 90 90

80 80

70 70 %

60 60 Febuxostat Febuxostat

Subjects, 50 n=3098 50 n=1636 Allopurinol Allopurinol 40 n=3092 40 n=1631

30 30 0 6 12 18 24 30 36 42 48 54 60 66 72 0 6 12 18 24 30 36 42 48 54 60 66 72 Visit, months Visit, months Gout Flares Over Time by Treatment Group Per CE‐14 Patient‐Year Exposure CARES

1.6

1.4

1.2 Febuxostat (overall population)

yr Allopurinol (overall population)

‐ 1.0 pt

0.8 per

0.6 Flares

0.4

0.2

0 0 to <6 6 to <12 12 to <18 18 to <24 24 to <30 30 to <36 36 to <42 42 to<48 48 to <54 54 to <60 60 to <66 66 to <72 Time interval, months Exposure, pt‐yr Febuxostat 1383 1169 977 823 705 598 510 416 344 266 195 129 Allopurinol 1377 1165 974 813 688 583 493 406 333 258 190 121

Modified intention‐to‐treat population (patients who were randomized to treatment and received at least one dose of study drug). CE‐15 Efficacy Conclusions

• Clinical studies have confirmed durable and dose‐dependent sUA reduction by febuxostat (40 mg and 80 mg) – Effectively lowers and maintains sUA <6 mg/dL • Febuxostat superior to commonly used doses of allopurinol – Effective in patients with mild/moderate renal impairment – Effective in patients with high sUA or tophi • Persistent sUA reduction demonstrated in long‐term studies, including CARES • Maintenance of sUA <6 mg/dL decreases gout flares and resolves tophi CR‐1

Benefit/Risk Assessment

John Affinito, MD Executive Medical Director Global Patient Safety & Evaluation, Marketed Products CR‐2 Benefit‐Risk Framework

• Analysis of condition • Current treatment options • Benefit assessment • Risk and risk management CR‐3 Analysis of Condition: Gout Is Serious, Chronic, and Debilitating

Urate deposition 3‐Dimensional Bone and joint within handsa CT imagea destructionb

a Dalbeth N, et al. Arthritis Rheumatol. 2007;56:29. b Image courtesy of Dr. Fernando Perez‐Ruiz. CR‐4 Limited Treatment Options

• Two xanthine oxidase inhibitors Allopurinol Febuxostat – Allopurinol or febuxostat Probenecid ± colchicine 1.5% • Two uricosuric agents 6.8% – Probenecid or lesinurad + XOI – Only probenecid used in clinical practice • One biologic agent – Pegloticase – Reserved for severe gout 91.7%

Proportion of prescriptionsa a IMS Health & NPA Monthly 2018. CR‐5 Benefit Assessment

• Reduces sUA to target, flares, and tophi in a simple one‐step titration • Beneficial in patients with renal impairment – No dose adjustment for patients with mild‐to‐moderate renal impairment – 40 mg for patients with severe renal impairment • Also, effective in severe gout (sUA ≥10 mg/dL or tophi) • No significant interactions with commonly used drugs • Important treatment option CR‐6 Risk Assessment

• Overall safety profile well‐established – 15 million patient‐years of exposure • CARES – Primary endpoint: febuxostat is noninferior to allopurinol regarding adjudicated MACE • Similar rates of non‐fatal CV events • Higher rate of CV death • Comprehensive assessment – CARES: Subgroup analyses, patient populations, and laboratory values/clinical data – Overall clinical program: Phase 3 trials, CV mechanistic studies, and TQT – Postmarketing surveillance and literature • Totality of data does not provide conclusive evidence that febuxostat is associated with an increased risk of CV death • Our goal is to inform prescribers so they can make informed treatment decisions for individual patients CR‐7 Scientific Exchange of CARES Data

• Shared top‐line results with the Agency (October 2017) – FDA issued a drug safety communication (November 2017) • Extensive scientific exchange with the medical community – The New England Journal of Medicine (March 2018)a – American College of Cardiology Congress (March 2018) – American College of Rheumatology Conference (October 2018)

aWhite WB, et al. N Engl J Med. 2018;378:1200‐1210. CR‐8 Communication to Prescribers and Continued Surveillance

• Communicate CARES data to prescribers – Update Warnings and Precautions • Include data on CARES and CV death • Carefully consider the risk and benefit of using febuxostat – Send Dear Healthcare Provider letter to prescribers, pharmacies, and professional societies • Continue postmarketing surveillance • Monitor the literature • Evaluate outcome of ongoing FAST CV safety study CR‐9 Summary and Conclusions

• Gout is a serious and debilitating disease with limited treatment options • CARES demonstrated: febuxostat was noninferior to allopurinol regarding adjudicated MACE. However, there was a higher rate of CV death • Robustness of the primary MACE is supported by subgroup and sensitivity analyses • It is important to communicate the CARES data to the medical community to enable informed and shared treatment decisions between providers and patients • Febuxostat remains an important treatment option CP‐1

Clinical Perspectives

N. Lawrence Edwards, MD Professor of Medicine University of Florida College of Medicine Gout: A Painful and Debilitating Disease With Limited CP‐2 Treatment Options

• Affects approximately 9 million people in the USa • Chronic condition – Increasingly frequent, painful, prolonged, and disabling flares; often in >1 joint – Development of tophi leading to bone and joint destruction and chronic disability – Compromised HRQOL and societal function – Increased risk for hospitalization – Increased mortality

a Chu‐Xu M, et al. Arthritis Rheumatol. 2019; In Press. CP‐3 Limited Urate‐Lowering Therapeutic Options in Gout With Many Treatment Challenges

XO Inhibitors Uricosuric Agents Pegloticase (pegylated recombinant modified Allopurinol and febuxostat Probenecid and lesinurad mammalian uricase) • Only ~50% pts reach sUA target • Both infrequently used • IV administration at commonly used doses • Probenecid • Indicated as therapy for • Allopurinol ‒ Requires multiple daily dosing progressive gout not responsive ‒ Challenges with multiple dose‐ ‒ Many drug‐drug interactions to conventional oral therapy titration steps to achieve target ‒ Limited efficacy in patients with • Development of high titers ‒ Under‐dosed in patients with impaired renal function of anti‐pegloticase antibody renal impairment • Lesinurad is associated with loss of urate‐ ‒ Rare, life‐threatening lowering efficacy and infusion hypersensitivity syndrome ‒ Approved only for combined reactions, including anaphylaxis therapy with an XOI • Febuxostat ‒ Not recommended in patients ‒ Cardiovascular warning for with CKD3 or higher thromboembolic events

Rashid N, et al. J Rheumatol. 2015;42(3):504‐512. CP‐4 Dose Titration of Xanthine Oxidase Inhibitors

Allopurinol Febuxostat CKD3 or better CKD4 Proportion of pts 800 achieving target sUA

80 67% 600 Proportion of pts achieving target sUA mg/day

400 50% 40 45%

Dose, 42% 200

0 012345678910 11121314 246 Dose escalations, months Weeks Khanna D, et al. Arth Care Res. 2012;64(10):1431‐1446; Perez‐Ruiz F, et al. Ann Rheum Dis. 1998;57(9):545‐549; Uloric [prescribing information]. Cambridge, MA: Takeda; 2018; Takada M, et al. J Clin Pharm Ther. 2005;30(4):407‐412. CP‐5 Flare Rate Reduction in Patients on Febuxostat

• Febuxostat decreases flares in patients with gout – This 5‐year study assessed urate‐lowering and clinical efficacy and safety of long‐term febuxostat therapy in subjects with gout. The primary efficacy endpoint was reduction to and maintenance of serum urate (sUA) levels <6.0 mg/dL

30 for 25 period

treated 20

month ‐ 2

15 subjects

of

previous

10 the

in 5 Percentage flare 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 n 27 27 19 14 7 12 4 4 5 5 2 4 3 2 3 1 1 1 1 3 1 2 1 3 1 0 1 0 0 0 N 116 96 89 88 84 83 76 72 71 69 69 69 68 66 65 65 64 62 61 60 59 56 55 55 55 55 55 55 54 54

Schumacher HR Jr, et al. Rheumatology (Oxford). 2009;48:188‐194. CP‐6 Overview of Benefits From Febuxostat

• Treating to target with XOIs reduces flares and leads to tophus resolution • Allopurinol has historically been underutilized – Less than 3% of all allopurinol prescriptions in the US are for doses > 300 mg/daya – 400 mg/day is the dose of allopurinol required to achieve target in ~50% of patients • Febuxostat can achieve sUA target in ~70% patients in one step • Advantages of febuxostat in patients with renal impairment – Renal excretion of febuxostat/metabolites is low – No need to dose reduce in mild or moderate renal impairment – Approved for patients with severe renal impairment at 40 mg • Febuxostat plays an important role in the management of gout and provides an effective alternative to allopurinol

a Sarawate CA, et al. Mayo Clin Proc. 2006;81(7):925‐934. Key Characteristics of Patients for Whom I Would CP‐7 Prescribe Febuxostat

• Allopurinol and febuxostat are first‐line therapies • Febuxostat when allopurinol needs to be discontinued – Lack of efficacy with optimal dose – GI distress – Transaminase elevation – Rash • Febuxostat as the first choice – Populations at risk for allopurinol hypersensitivity syndrome (AHS) • Southeast Asians, Han Chinese, Koreans – Inadequate sUA lowering with optimal allopurinol therapy – Resistance from primary care physicians or nephrologist – In patients where the allopurinol dose‐escalation scheme is overwhelming • Before prescribing ULT, I discuss risk/benefits, including CARES trial findings CP‐8 Conclusions

• Strong evidence supports the role of xanthine oxidase inhibitors in treating all clinical aspects of gouty arthritis • Based on the totality of evidence, febuxostat has a positive benefit/risk profile • Takeda is recommending appropriate risk‐management steps to address the observation of a higher rate of CV deaths in the CARES trial • Febuxostat should continue to play an important role in the treatment of gout BC-15 Discontinuation From Study by Country CARES

Febuxostat Allopurinol Country n=3098 n=3092 United States 1081/2651 (40.8%) 1102/2665 (41.4%)

Canada 20/68 (29.4%) 13/72 (18.1%)

Mexico 48/379 (12.7%) 53/355 (14.9%) NC-29 No Risk of Cross Hypersensitivity Between Allopurinol and Febuxostat

• Cohort study of patients from Taiwan receiving allopurinol or febuxostat1 ― Patients with allopurinol hypersensitivity did not exhibit hypersensitivity to febuxostat ― In Asians, incidence and severity of hypersensitivity reactions related to febuxostat are less than that of allopurinol d Febuxostat Allopurinol New users (N) 5,680 12,007 Hypersensitivity reactions identified 1 (0.2 / 1000 new users) 33 (2.7 / 1000 new users) Adverse reactions to subsequent Febuxostat in 0 (0/14) patients with allopurinol hypersensitivity

• Patients with allopurinol-SCARa showed no cross reactivity to febuxostat2 ― T- cells from patients with allopurinol–SCAR did not cross react to febuxostat in a granulysin-based T- cell activation assayb ― 10 patients with allopurinol-SCAR (all 10 HLA-B*58:01) were prescribed febuxostat and none developed recurrence of SCAR a SCAR – Severe Cutaneous Adverse Reactions. b Study demonstrated that a granulysin-based T-cell activation assay is a sensitive method for examining the causative drug of allopurinol SCAR 1Chen C, Value in Health 2018; 21: S106, 2Chung W, et al. The Society for Investigative Dermatology 2015; 00: 1-12. SS-4 CV Death by Timing of Event CARES

Subjects, n (%) Febuxostat Allopurinol Hazard ratioa n=3098 n=3092 (95% CI) mITT 134 (4.3) 100 (3.2) 1.34 (1.03, 1.73)

On drug 23 (0.7) 14 (0.5) 1.62 (0.84, 3.15)

After last dose 111 (3.6) 86 (2.8) 1.28 (0.97, 1.70)

On drug + 30 days 62 (2.0) 41 (1.3) 1.49 (1.01, 2.22)

On drug + 60 days 76 (2.5) 53 (1.7) 1.42 (1.00, 2.01)

On drug + 120 days 92 (3.0) 73 (2.4) 1.24 (0.92, 1.69)

On drug + 180 days 107 (3.5) 85 (2.7) 1.24 (0.94, 1.65)

a Febuxostat to allopurinol. SS-16 CV Deaths Within 30 Days After Discontinuing Febuxostat AEs (Verbatim Term) With Timing of Last Dose and Death

Cardiac arrest AE leading to death Cardiac arrest Death Pulmonary fibrosis Cardiovascular shock Septic shock Congestive heart failure Worsening of congestive heart failure Acute MI Ventricular fibrillation Hypotension Non-ST elevated MI 2/2 type II NSTEMI Intracranial hemorrhage End-stage ischemic cardiomyopathy Cerebral infarction End-stage congestive heart failure Respiratory failure Cerebral vascular accident Cardiac arrest Cardiac arrest Suspected MI Death (unknown cause) Ischemic heart disease Sudden cardiac death Pulseless electrical activity arrest Cardiogenic shock MI MI Cardiac arrhythmia Acute cardiopulmonary collapse Death due to undetermined natural disease Acute fatal MI Congestive heart failure Unspecified cardiac arrest Natural end-stage coronary artery disease Cardiogenic shock Sudden death Heart attack Heart attack

-Y1 Last dose Day 30 CSR Figure 11.b SS-21 CV Death by Country CARES

Subjects, n (%) Country Febuxostat Allopurinol United States 125/2651 (4.7) 90/2665 (3.4) Canada 2/68 (2.9) 2/72 (2.8) Mexico 7/379 (1.8) 8/355 (2.3) Exploratory Endpoint - Time to First Occurrence of Expanded SS-30 Cardiovascular Endpoint (mITT Population) CARES

Febuxostat Allopurinol Favors Endpoint n=3098 n=3092  Febuxostat Allopurinol  P value Expanded endpointa 560 (18.1) 525 (17.0) 0.353 CV death 134 (4.3) 100 (3.2) 0.027 Nonfatal MI 111 (3.6) 118 (3.8) 0.610 Nonfatal stroke 71 (2.3) 70 (2.3) 0.944 Urgent revascularization due to unstable angina 49 (1.6) 56 (1.8) 0.445 Urgent cerebral revascularization (non-elective) 0 1 (<0.1) Not calculated Hospitalized heart failure 134 (4.3) 121 (3.9) 0.462 Arrhythmias not associated with ischemia 112 (3.6) 113 (3.7) 0.913 Venous thromboembolic events 58 (1.9) 45 (1.5) 0.201 Transient ischemic attack, hospitalized 28 (0.9) 23 (0.7) 0.523 0.25 0.5 1 2 4 Hazard ratio (95% CI) a CV death, nonfatal MI, nonfatal stroke, urgent revascularization due to unstable angina, urgent cerebral revascularization, hospitalized heart failure, arrhythmias not associated with ischemia, venous thromboembolic events, transient ischemic attacks. SS-40 Episode of Gout Flare Within 3 Months Prior to Event CARES

Subjects, n (%) Febuxostat Allopurinol Event n=3098 n=3092 Experienced gout flare within 3 months of event? CV death 134 100 Yes 8 (6.0) 2 (2.0) No 126 (94.0) 98 (98.0) Non-fatal MI 111 118 Yes 16 (14.4) 7 (5.9) No 95 (85.6) 111 (94.1) Non-fatal stroke 71 70 Yes 5 (7.0) 6 (8.6) No 66 (93.0) 64 (91.4) Unstable angina 49 56 Yes 9 (18.4) 9 (16.1) No 40 (81.6) 47 (83.9) Risk Ratios (95% CI) for Primary MACE Composite Endpoint SS-54 by Subgroups CARES

Interaction Baseline variables  Favors febuxostat Favors allopurinol  RR (95% CI) P value Age, yr <65 0.97 (0.77, 1.23) 0.380 ≥65 1.11 (0.92, 1.33) Gender Female 0.94 (0.63, 1.41) 0.607 Male 1.06 (0.90, 1.23) BMI, kg/m2 <30 1.06 (0.82, 1.36) 0.886 ≥30 1.03 (0.86, 1.23) Race White 1.02 (0.87, 1.20) 0.652 Non-white 1.11 (0.80, 1.56) Years since gout diagnosis <5 1.05 (0.83, 1.32) 0.461 5-10 0.85 (0.59, 1.22) >10 1.11 (0.89, 1.37) Baseline sUA, mg/dL <9.0 1.00 (0.81, 1.22) 0.779 9.0-10.0 1.13 (0.84, 1.53) ≥10.0 1.03 (0.78, 1.36) NSAID use Yes 1.25 (0.97, 1.62) 0.096 No 0.96 (0.81, 1.15) Low-dose aspirin use Yes 0.92 (0.75, 1.13) 0.091 No 1.18 (0.96, 1.46) Colchicine used during study Yes 1.23 (0.94, 1.61) 0.149 No 0.97 (0.82, 1.16) 0.5 1 2 Risk ratio (95% CI) SS-64 Risk Ratios for CV Death by NSAID Use CARES

Subjects, % (n/N) Febuxostat Allopurinol Lower risk on Risk ratio (95% CI) NSAID use n=3098 n=3092  Febuxostat Allopurinol  (febuxostat vs allopurinol) P value At baseline (start date prior to randomization) Yes 4.8 (41/856) 2.2 (20/908) 2.17 (1.28, 3.68) 0.032 No 4.1 (93/2242) 3.7 (80/2184) 1.13 (0.84, 1.52) During treatment (use overlaps with DB period) Yes 3.3 (26/779) 2.5 (20/798) 1.33 (0.75, 2.37) 0.985 No 4.7 (108/2319) 3.5 (80/2294) 1.34 (1.01, 1.77)

0.5 1 2 4 Risk ratio (95% CI) SS-65 Risk Ratios for CV Death by Aspirin Use CARES

Subjects, % (n/N) Febuxostat Allopurinol Lower risk on Risk ratio (95% CI) Aspirin use n=3098 n=3092  Febuxostat Allopurinol  (febuxostat vs allopurinol) P value At baseline (start date prior to randomization) Low dose (<325 mg) Yes 3.3 (49/1496) 3.8 (57/1481) 0.85 (0.58, 1.24) 0.001 No or at higher doses 5.3 (85/1602) 2.7 (43/1611) 1.99 (1.39, 2.85) Any dose Yes 4.0 (76/1895) 3.7 (72/1933) 1.08 (0.78, 1.48) 0.026 No 4.8 (58/1203) 2.4 (28/1159) 2.00 (1.28, 3.11) During treatment (use overlaps with DB period) Low dose (<325 mg) Yes 3.4 (57/1660) 3.8 (63/1645) 0.90 (0.63, 1.27) 0.001 No or at higher doses 5.4 (77/1438) 2.6 (37/1447) 2.09 (1.42, 3.08) Any dose Yes 4.0 (80/2017) 3.5 (72/2039) 1.12 (0.82, 1.54) 0.064 No 5.0 (54/1081) 2.7 (28/1053) 1.88 (1.20, 2.94)

0.5 1 2 4 Risk ratio (95% CI) SS-68 Risk Ratio for CV Death by History of Urate-Lowering Therapy CARES

RR (95% CI) Febuxostat Allopurinol Favors (Febuxostat vs History of ULT n=3098 n=3092  Febuxostat Allopurinol  Allopurinol) P value

Yes 4.6 (88/1914) 3.6 (68/1914) 1.29 (0.95, 1.76) 0.932

No 3.6 (38/1045) 2.9 (30/1044) 1.27 (0.79, 2.03)

0.5 1 2 4 Risk ratio (95% CI) SS-101 Adjudicated APTC CV Events by Subject-Year Long-Term Extension Studies

Subjects (rate per 100 subject-yr) Febuxostat 40 mg 80 mg 120 mg Total Allopurinol n=12 n=917 n=524 n=1143 n=178 PY=37.7 PY=1745.6 PY=877.7 PY=2660.9 PY=172.2 All APTC events 1 (2.66) 17 (0.97) 9 (1.03) 27 (1.01) 1 (0.58) 0.07, 14.79 0.57, 1.56 0.47, 1.95 0.67, 1.48 0.02, 3.24 CV death 0 4 (0.23) 3 (0.34) 7 (0.26) 0 0.00, 9.79 0.06, 0.59 0.07, 1.00 0.11, 0.54 0.00, 2.14 Nonfatal MI 0 8 (0.46) 3 (0.34) 11 (0.41) 1 (0.58) 0.00, 9.79 0.20, 0.90 0.07, 1.00 0.21, 0.74 0.02, 3.24 Nonfatal stroke 1 (2.66) 5 (0.29) 3 (0.34) 9 (0.34) 0 0.07, 14.79 0.09, 0.67 0.07, 1.00 0.16, 0.64 0.00, 2.14

PY=subject-year. SS-116 No Effect of Febuxostat on Platelet Aggregation in the Presence of ADP ST-3 Complete and Partial Follow-up CARES

Subjects, n (%) Follow-up vs Status at study Median study Febuxostat Allopurinol Total primary endpoint close out follow-up n=3098 n=3092 n=6190

Complete MACE event 2.0 years 335 (10.8) 321 (10.4) 656 (10.6)

Complete Non-MACE Death 2.7 years 105 (3.4) 92 (3.0) 197 (3.2)

Complete Censored at close out 3.7 years 1676 (54.1) 1677 (54.2) 3353 (54.2)

Partial No visit at close out 1.7 years 982 (31.7) 1002 (32.4) 1984 (32.1)

Of those with partial follow, approx. 6.5% might have had a MACE event ST-5 Additional Events to Result in Failure to Show MACE Noninferiority CARES

Subjects, n (%) Febuxostat Allopurinol Hazard Ratio Risk difference n=3098 n=3092 (95% CI) Per 100 pt-yr (95% CI) P value Composite primary endpoint 335 (10.8) 321 (10.4) 1.03 0.12 0.66 (0.87, 1.23) (-0.55, 0.79)

Expected number of HR amongst New New Risk in missed events MACE events missed missed events overall HR upper CL vs primary result 6.5% 1.65 1.130 1.299 22 fold higher risk 2 x 6.5% 1.39 1.140 1.299 13 fold higher risk Additional Events to Result in Significant Increase in the Risk of ST-7 All Cause Death With Additional Vital Status Data CARES Subjects, n (%) Febuxostat Allopurinol Hazard Ratio n=3098 n=3092 (95% CI) P value All-cause mortality with 332 1.09 309 (10.0) 0.25 additional vital status data (10.7) (0.94, 1.28)

Expected number of HR amongst New Risk in missed events events missed missed events overall HR P-value vs primary result 5.7% 3.1 1.154 <0.05 23 fold higher risk 2 x 5.7% 1.83 1.156 <0.05 9 fold higher risk