The Resurrection of Ezetimibe: A Story of Perseverance

Transcription

1 The Resurrection of Ezetimibe: A Story of Perseverance Harry Chip Davis CVPath Institute Gaithersburg, MD (Schering-Plough / Merck) 1

2 Ezetimibe Inhibits Intestinal Cholesterol Uptake and Absorption (Discovered in mid 1990s by in-vivo cholesterol absorption screening) 1000 mg ABCG5/G8 MTP Transporter or Passive Diffusion? Ezetimibe?

3 Ezetimibe and Its Glucuronidated Metabolite (SCH 60663) OH OH OH O HO O OH OH CO 2 H F O N Glucuronidation F UGT1A1, O N Ezetimibe F UGT1A3, UGT2B15 SCH (Glucuronide) F Ezetimibe and its glucuronide metabolite are potent inhibitors of dietary and biliary cholesterol absorption

4 % Fractional Cholesterol Absorption (Geometric Mean) Reduction of Cholesterol Absorption in Humans With Ezetimibe (10mg per day 2 week crossover study) Ezetimibe Inhibited Cholesterol Absorption by 54% %* % 22.7% 10 0 Placebo Ezetimibe *P<0.001

5 Ezetimibe had No Target Organ Toxicity in Mice, Rats, and Dogs and No Evidence for Carcinogenicity in 2-Year Oncogenicity Studies in Mice and Rats (Halleck, Davis et al. Toxicology 258: , 2009)

6 Mean % change in LDL from baseline at wk 12 Ezetimibe Monotherapy (Approved 2002) Maximal Efficacy at 10 mg/day Phase IIb Monotherapy Dose-Response Study: LDL-C Reduction at Week Placebo (n=52) +4.3 Ezetimibe 0.25 mg (n=47) Ezetimibe 1 mg (n=49) Ezetimibe 5 mg (n=49) Ezetimibe 10 mg (n=46) * * * * -30 Bays et al. Clin.Ther. 2001;23: * p<.01 vs placebo

7 Plasma Cholesterol (mg/dl) Ezetimibe Prevents Diet-Induced Hypercholesterolemia in Monkeys: Why doesn t ezetimibe reduce plasma cholesterol below baseline levels? Compensatory increase in cholesterol synthesis. 350 Control Ezetimibe (0.1 mg/kg/d) Crossover 20 Time (days) van Heek, Davis et al. Eur J Pharm. 2001; 415:79-84

8 Ezetimibe + Lovastatin Lowers Plasma Cholesterol in Chow-Fed Dogs in a Synergistic Manner Change in Plasma Cholesterol (mg/dl) 40 Control (n=5) Ezetimibe, mg/kg (n=5) Lovastatin, 5 mg/kg (n=5) Combination (n=5) ± ±16 136± ± Days of Treatment Davis, Watkins, et al. Metabolism 2001; 50:

9 Dual Inhibition of Cholesterol Synthesis and Absorption Diet Bile Bile acids Statin Liver Small intestine Absorption Cholesterol Synthesis Ezetimibe Excretion in feces Atherogenic lipoproteins 9

10 Ezetimibe Combined with Simvastatin (Vytorin / Inegy Approved 2004) Reduces LDL Cholesterol and C-Reactive Protein LDL-Cholesterol CRP Pearson T et al, Am J Cardiol 99: 1706,

11 Ezetimibe: Clinical Summary Selective intestinal cholesterol (10 mg average 54% to 65% reduction) and phytosterol absorption inhibitor (no effect on fat soluble vitamins, triglycerides, etc) Maximally effective dose is 10 mg once a day for hypercholesterolemia and sitosterolemia Extremely low systemic plasma levels (Cmax um), circulates enterohepatically, repeatedly delivering the agent back to the site of action (glucuronide metabolite 88% of total plasma drug concentration) LDL-C reduction 18% as monotherapy, and incremental 18-25% additional reduction with statin co-administration Plasma phytosterol levels reduced approximately 50% in patients with sitosterolemia or hypercholesterolemia

12 Molecular Target of Ezetimibe (Zetia, Ezetrol)? Intestinal Localization 125 I-Gluc-Ezetimibe in Rats

13 Genomic/ Bioinformatic Strategy to Identify the Molecular Target of Ezetimibe All known sterol processing mechanisms, as well as biochemical and molecular approaches were evaluated for more than 12 years. All identified proteins were not involved in the mechanism of action of Ezetimibe. Genomic/ Bioinformatic Strategy Hypothetical properties of an ideal candidate cholesterol transporter: Will be expressed in enterocytes in the proximal small intestine restricted tissue expression Membrane bound protein expressed on the surface of the cell Contain features found in proteins involved in sterol metabolism e.g. cholesterol regulated expression, sterol sensing domain Created and sequenced a rat small intestine DNA library of > 16,000 genes Using bioinformatics these sequences were assembled and human and mouse databases mined for the properties of the ideal cholesterol transporter

14 Niemann-Pick C1 Like 1 Protein Is Critical for Intestinal Cholesterol Absorption Scott W. Altmann,* Harry R. Davis, Jr., Li-ji Zhu, Xiaorui Yao, Lizbeth M. Hoos, Glen Tetzloff, Sai Prasad N. Iyer, Maureen Maguire, Andrei Golovko, Ming Zeng, Luquan Wang, Nicholas Murgolo, Michael P. Graziano Dietary cholesterol consumption and intestinal cholesterol absorption contribute to plasma cholesterol levels, a risk factor for coronary heart disease. The molecular mechanism of sterol uptake from the lumen of the small intestine is poorly defined. We show that Niemann-Pick C1Like 1(NPC1L1) protein plays a critical role in the absorption of intestinal cholesterol. NPC1L1 expression is enriched in the small intestine and is in the brush border membrane of enterocytes. Although otherwise phenotypically normal, NPC1L1-deficient mice exhibit a substantial reduction in absorbed cholesterol, which is unaffected by dietary supplementation of bile acids. Ezetimibe, a drug that inhibits cholesterol absorption, had no effect in NPC1L1 knockout mice, suggesting that NPC1L1 resides in an ezetimibe-sensitive pathway responsible for intestinal cholesterol absorption. Science 303: , Feb 20, 2004

15 Human NPC1L1 Loop with Sterol Binding Pocket Coding SNPs (Red) Ezetimibe Binding Site I647N Davis HR et al Current Opinion in Lipidology 22: , 2011

16 Ezetimibe Prevents the Internalization of NPC1L1 and Uptake of Free Cholesterol In-Vivo Xie C, Song B-L, et al J Lipid Res 53:2092, 2012

17 Model of NPC1L1/Cholesterol Uptake (P-S Li, B-L Song, et al Nature Medicine 20:80-86, 2014)

18 Intestinal Specific Cholesterol Absorption Inhibitors AZD4121 [45] F OH S O N O O N H H N O O OH AstraZeneca Human recombinant?=0.065 Mouse=0.08 um/kg/? Phase III Discontinued, LDL-C 12mg F Canosimibe (AVE5530) [46] F OH O N O CH3 H N O O N H OH OH OH OH OH Sanofi-aventis NA Hamster=Slightly more potent than ezetimibe Phase III, 12 weeks Discontinued LDL-C 25 mg KT6-971[47] OH HO HO O OH OH Kotobuki Pharmaceutical Human recombinant=0.99 Hamster=0.23 mg/kg/7d Phase I discontinued F O N MD-0727 [48] (A) Microbia pharmaceuticals NA Rat=68%@0.3mg/kg/3h Mouse=76%@3mg/kg/ 3h Hamster=88%@1mg/kg /4h Phase I discontinued (B) MK-6213 [49] OH OH OH OH Merck Human recombinant =0.005 Mouse=28ug/kg/5h Rat =3.3ug/kg/16h Hamster=68ug/kg/7d Phase 2 Discontinued, LDL-C %@160mg F O N H N S CH 3 O O Ezetimibe 307nM Ez-Gluc 150nM Definitive structure was not available for MD Patents reference novel quaternary salt derivatives (A) and tethered dimers and trimers of 1,4 diphenylacetidin-2-ones (B). Davis HR et al Current Opinion in Lipidology 22: , 2011

19 Hepatic Human NPC1L1 Transgenic Mice have Reduced Biliary Cholesterol and Increased Plasma Cholesterol, which is Normalized by Ezetimibe Bile Temel et al, JCI 117:1968, 2007

20 BY BLOCKING CHOLESTEROL UPTAKE IN THE LIVER AND INTESTINE, EZETIMIBE FOSTERS GREATER ELIMINATION OF FREE CHOLESTEROL 25% Dietary Chol Intestine Cholesterol Pool (Micelles) Cholesterol and plant sterols Ezetimibe Cholesterol and plant sterols NPC1L1 ABCG5/G8 75% Biliary chol Enterocyte Ezetimibe ACAT MTP Free Chol Chol ester HMGCoA Red Acetyl CoA Statin NPC1L1 Free Chol ABCG5/G8 Bile Acids Remnant receptors CM Acetyl CoA Chol Cholesterol Pool CMR Lipases TG Liver HMGCoA Red CE CETP TG TG SR-BI Statin HDL (mature) LCAT LDLR CE CETP TG HDL (nascent) LDL Hepatic ApoB-100 VLDL VLDL-R/ ldl TG Lipases TG Peripheral Tissues Fecal sterols Intestinal Apo B-48 macrophage Blood Atheroma Davis HR et al Current Opinion in Lipidology 22: , 2011

21 Ezetimibe Inhibits Atherosclerosis in Animal Models of Atherosclerosis Davis HR, et al. Atherosclerosis 2011;215:

22 Innominate Arteries of 8 month old apoe null and 2 year old Npc1l1/apoE null Male Mice Fed a Chow Diet ApoE null 8 month old Npc1l1/ApoE null 2 year old - 78% Davis HR et al ATVB 27: , 2007 and Unpublished

23 Inactivating Mutations in NPC1L1 and Protection from Coronary Heart Disease Naturally occurring mutations that disrupt NPC1L1 function were found to be associated with reduced plasma LDL cholesterol levels (12mg/dl) and a reduced risk of coronary heart disease (53%) (heterozygous). The Myocardial Infarction Genetics Consortium Investigators. N Engl J Med 2014;371: Sekar Kathiresan

24 Ezetimibe/Simvastatin Surrogate and Clinical Outcome Program Trial Population Endpoint Treatment ENHANCE: Started: 6/02 HeFH (n = 720) Carotid IMT EZE/Simva 10/80 vs Simva 80 SEAS Started: 3/03 Aortic Stenosis (n = 1873) MACE EZE/Simva 10/40 vs PBO SHARP Started: 7/03 CKD (n = 9438) MACE EZE/Simva 10/20 vs PBO vs Simva 20 IMPROVE-IT Started: 10/05 End: 2014 ACS (n = 18,000) MACE EZE/Simva 10/40 vs Simva 40

25 Percentage change from baseline ENHANCE: LDL-cholesterol (HeFH) ENHANCE P< Months Baseline 24 months (mg/dl) (mg/dl) Simva 318 ± ± 60 Eze-Simva 319 ± ± % incremental reduction Simva Eze-Simva

26 cimt mm ENHANCE: Critical Factors for Successful cimt Trial progression ASAP ASAP Simva LDLc -40% 0.85 Atorva LDLc -52% 0.80 regression ENHANCE Simva No progress; Sim/Eze No regress FAILURE? P= ns ENHANCE Simva LDLc -40% Simva/Eze LDLc -57% years Stein: ENHANCE failed drug or failed trial

27 Mean (mg/dl) SEAS Trial LDL-Cholesterol Intention to Treat Population 150 Placebo EZ/Simva 10/40 mg Year

28 Patients (%) with First Event Ischemic Cardiovascular Events 30 (ICE) Outcome in SEAS Intent-to-Treat Population 20 Placebo 10 Hazard ratio: 0.78, P = E/S (10/40 mg) 0 # patients at risk E/S Placebo Years in Study Rossebø AB et al. N Engl J Med. 2008;359:

29 In-Depth Assessment of Cancer Data An Assessment Of The Carcinogenic Potential Of Ezetimibe Using Nonclinical Data In A Weight-of-evidence Approach M. Halleck, H.R. Davis, P. Kirschmeier,D.Levitan, R.D. Snyder, K. Treinen and J.S. MacDonald In conclusion, the non-clinical data do not support the proposed hypothesis based on the single observation from the SEAS trial and, rather, support the conclusion that ezetimibe does not represent a carcinogenic hazard to humans using this drug in a therapeutic setting." Toxicology 258: , 2009

30 The SHARP Study Study of Heart and Renal Protection VYTORIN (ezetimibe/simvastatin) and ZETIA (ezetimibe) in Patients with Chronic Kidney Disease SHARP was designed and implemented by the Clinical Trial Service Unit (CTSU) at the University of Oxford Baigent C. et al. Lancet 377: ,

31 LDL-C difference (mg/dl) SHARP Trial Effect on LDL-cholesterol (LDL-C) at 1 year of three-quarters compliance with eze/simva 0 eze/simva vs simva simva vs placebo eze/simva vs placebo SHARP Collaborative Group Am Heart J 2010

32 Proportion suffering event (%) 25 SHARP (Study of Heart and Renal Protection) Key outcome: Major Atherosclerotic Events Risk ratio 0.83 ( ) Logrank 2P= placebo eze/simva 5 0 Baigent C. et al. Lancet 377: , Years of follow-up

33 IMProved Reduction of Outcomes: Vytorin Efficacy International Trial A Multicenter, Double-Blind, Randomized Study to Establish the Clinical Benefit and Safety of Vytorin (Ezetimibe/Simvastatin Tablet) vs Simvastatin Monotherapy in High-Risk Subjects Presenting With Acute Coronary Syndrome

34 Study Design Patients stabilized post ACS 10 days: LDL-C *mg/dL (or **mg/dL if prior lipid-lowering Rx) *3.2mM **2.6mM N=18,144 Standard Medical & Interventional Therapy Simvastatin 40 mg Uptitrated to Simva 80 mg if LDL-C > 79 (adapted per FDA label 2011) Ezetimibe / Simvastatin 10 / 40 mg Follow-up Visit Day 30, every 4 months 90% power to detect ~9% difference Duration: Minimum 2 ½-year follow-up (at least 5250 events) Primary Endpoint: CV death, MI, hospital admission for UA, coronary revascularization ( 30 days after randomization), or stroke Cannon CP AHJ 2008;156:826-32; Califf RM NEJM 2009;361:712-7; Blazing MA AHJ 2014;168:205-12

35 Milestones and Events SEAS Study Results (Jul-2008) FDA Simva Label Change 80mg dose (Jun-2011) FPI 26-Oct-2005 ENHANCE Study Results (Jan-2008) ARBITER-6 (Nov-2009) LPI 08-Jul Events Sep 2007: increase n=12,500 and events to 5250, cap STEMI April 2008: increase n=18,000 50% Interim Analysis Mar-2010 DSMB Review (88%) Mar % Interim Analysis Mar-2012 Database Lock Oct-2014

36 Study Metrics Simva (N=9077) EZ/Simva (N=9067) Uptitration to Simva 80mg, % 27 6 Premature study drug D/C, % Median follow-up, yrs Withdraw consent w/o vital status, %/yr Lost to follow-up, %/yr Follow up for primary endpoint, % Follow up for survival, % Total primary endpoint events = 5314 Total patient-years clinical follow-up = 97,822 Total patient-years follow-up for survival = 104,135

37 Participant Disposition for 1 Endpoint OT Population *1 event on drug (4011) or non-cv death on drug or full assessment on drug during closeout Completed on drug* ITT: 18,144 ( EPs*) OT: 17,706 *EPs = endpoints 438 Drug never taken Off drug 30 days n = 10,573 n = 7133 Mean years of F/U on drug for S yrs. EZ/S 5292 S yrs. EZ/S 3559 Followed off drug ( ITT EPs* occurred > 30d off drug) primary endpoint = 60,298 total patient-years of F/U OT 80,286 patient years follow up for primary endpoint in ITT

38 Baseline Characteristics Simvastatin (N=9077) % EZ/Simva (N=9067) % Age (years) Female Diabetes MI prior to index ACS STEMI / NSTEMI / UA 29 / 47 / / 47 / 24 Days post ACS to rand (IQR) 5 (3, 8) 5 (3, 8) Cath / PCI for ACS event 88 / / 70 Prior lipid Rx LDL-C at ACS event (mg/dl, IQR) 95 (79, 110) 95 (79,110)

39 LDL-C and Lipid Changes 1 Yr Mean LDL-C TC TG HDL hscrp Simva EZ/Simva Δ in mg/dl Median Time avg 69.5 vs mg/dl

40 Mean LDL-C at 1 Year OT & ITT LDLC values at 1 year Simva OT LDLC 69.5 mg/dl EZ/Simva OT LDLC 52.5 mg/dl OT LDLC 17.0 mg/dl Simva ITT LDLC 69.9 mg/dl EZ/Simva ITT LDLC 53.2 mg/dl ITT LDLC 16.7 mg/dl Simva ITT OT EZ/Simva ITT OT

41 Primary Endpoint ITT Cardiovascular death, MI, documented unstable angina requiring rehospitalization, coronary revascularization ( 30 days), or stroke HR CI (0.887, 0.988) p=0.016 Simva 34.7% 2742 events NNT= 50 EZ/Simva 32.7% 2572 events 6.4% Treatment effect Reaffirms the LDL-C hypothesis 7-year event rates

42 Primary Endpoint On-Treatment HR CI (0.868, 0.983) p=0.012 Simva KM 32.4% 2079 events EZ/Simva KM 29.8% 1932 events 7.6% Treatment effect Primary Endpoint: CV death, MI, hospital admission for UA, coronary revascularization (> 30 days after randomization), or stroke

43 Primary Endpoint On Treatment HR CI (0.868, 0.983) p=0.012 Simva KM 32.4% 2079 events NNT =38 EZ/Simva KM 29.8% 1932 events 19% greater treatment effect than ITT Primary Endpoint: CV death, MI, hospital admission for UA, coronary revascularization (> 30 days after randomization), or stroke 7 year event rates

44 Primary and 3 Prespecified Secondary Endpoints ITT & OT Simva* EZ/Simva* HR Primary ITT On Treatment Secondary ITT #1 OT Secondary ITT #2 OT Secondary ITT #3 OT Ezetimibe/Simva 1.0 Simva 1.1 Better Better *7-year event rates 1 = All-cause death, major coronary event, or stroke post randomization 2 = CHD death, non-fatal MI, or urgent CABG or PCI (>30 days) after randomization 3 = CV death, non-fatal MI, documented UA requiring rehospitalization, all revascularization (>30 days) after randomization, or non fatal stroke

45 Individual Cardiovascular Endpoints and CVD/MI/Stroke HR Simva* EZ/Simva* p-value All-cause death CVD CHD MI Stroke Ischemic stroke Cor revasc 30d UA CVD/MI/stroke Ezetimibe/Simva Better Simva Better *7-year event rates (%)

46 Major Pre-specified Subgroups Simva EZ/Simva Male Female Age < 65 years Age 65 years No diabetes Diabetes * Prior LLT No prior LLT LDL-C > 95 mg/dl LDL-C 95 mg/dl Ezetimibe/Simva Better Simva Better 7-year event rates *p-interaction = 0.023, otherwise > 0.05

47 IMPROVE-IT vs. CTT: Ezetimibe vs. Statin Benefit IMPROVE-IT CTT Collaboration. Lancet 2005; 366: ; Lancet 2010;376:

48 Safety ITT No statistically significant differences in cancer or muscle- or gallbladder-related events Simva n=9077 % EZ/Simva n=9067 % p ALT and/or AST 3x ULN Cholecystectomy Gallbladder-related AEs Rhabdomyolysis* Myopathy* Rhabdo, myopathy, myalgia with CK elevation* Cancer* (7-yr KM %) * Adjudicated by Clinical Events Committee % = n/n for the trial duration

49 Conclusions IMPROVE-IT: First trial demonstrating incremental clinical benefit when adding a non-statin agent (ezetimibe) to statin therapy: YES: Non-statin lowering LDL-C with ezetimibe reduces cardiovascular events YES: Even Lower is Even Better (achieved mean LDL-C 53 vs. 70 mg/dl at 1 year) YES: Confirms ezetimibe safety profile Reaffirms the LDL hypothesis, that reducing LDL-C prevents cardiovascular events Results could be considered for future guidelines