Drug Class Review Drug Effectiveness Review Project Marian McDonagh, PharmD Principal Investigator Statins and Fixed-dose Combination Products Containing a Statin Final Report Update 5 December 2009 M.E. Beth Smith, DO Nancy J. Lee, PharmD, BCPS Elizabeth Haney, MD Susan Carson, MPH

2 Populations –Outpatients targeted for primary or secondary prevention of coronary heart disease of non- coronary forms of atherosclerotic disease with or without hypercholesteremia –Inpatients with acute coronary syndrome or undergoing revascularization –Children: New population added this update Inclusion criteria

3 Study designs (no changes this update) –For effectiveness: Controlled clinical trials and good-quality systematic reviews –For efficacy (lipid levels): Head-to-head trials only –For adverse effects: Controlled clinical trials and observational studies

Interventions Individual statins –Atorvastatin (Lipitor ® ) –Fluvastatin (Lescol ® ) –Fluvastatin extended release (Lescol ® Xl) –Lovastatin (Mevacor ® ) –Lovastatin extended release (Altoprev ® ) –Pravastatin (Pravachol ® ) –Rosuvastatin (Crestor ® ) –Simvastatin (Zocor ® ) Fixed-dose combination products containing a statin (added this update) –Lovastatin; niacin extended release (Advicor ® ) –Simvastatin; ezetimibe (Vytorin ® ) –Simvastatin; niacin extended release (Simcor ® ) 4

5 Outcomes No changes this update Effectiveness outcomes –Reduction in nonfatal MI, CHD, mortality (CHD and all-cause, stroke, and need for revascularization (including coronary artery bypass grafting, angioplasty and coronary stents) Efficacy outcomes – LDL cholesterol lowering ability –HDL cholesterol raising ability –Percentage of patients meeting lipid goals Harms –Overall adverse events –Withdrawals due to adverse events –Serious adverse events –Specific adverse events (including, but not limited to, hepatotoxicity, myopathy, rhabdomyolysis, renal toxicity, myalgia)

6 Bibliographic databases –End date: June 2009 –Sources: Cochrane Library (CCRCT, CDSR, DARE), Medline Pharmaceutical company submissions (Update 5) –Rosuvastatin, fluvastatin –Simvastatin/niacin and simvastatin/ezetimibe fixed-dose combination products Reference lists US Food and Drug Administration reviews Literature searches

(3089 new with this update) total citations (2751) excluded at title/abstract level 791(338) full-text articles retrieved 444 (203) articles excluded at full-text level 347 (135) included studies 102 (24) head-to-head trials 29 (25) active-control trials 2 (1) head-to-head and active control trials 92 (24) placebo-controlled trials 80 (38) observational studies 21 (8) systematic reviews 21 (15) other Search results

Results: Adults 8

Key Question 1a How do statins and fixed-dose combination products containing a statin and another lipid lowering drug compare in their ability to reduce LDL cholesterol? Are there doses for each statin or fixed-dose combination product containing a statin and another lipid lowering drug that produce similar percent reduction in LDL cholesterol between statins? 9

Key Question 1a: Previous conclusions For patients who require LDL cholesterol reductions of up to 35% to meet their goal, any of the statins are effective. In patients requiring an LDL cholesterol reduction of 35% to 50% to meet the National Cholesterol Education Program goal, atorvastatin 20 mg or more, lovastatin 80 mg, rosuvastatin 10 mg or more, and simvastatin 20 mg daily are likely to meet the goal. Among high-potency/high-dose statins: –Atorvastatin 80 mg daily and rosuvastatin 20 mg or more reduced LDL cholesterol by 50% or more. –Atorvastatin 80 mg had a higher rate of some adverse effects (gastrointestinal disturbances and transaminase elevation) than simvastatin 80 mg daily in a trial in which the LDL lowering of atorvastatin was greater than that of simvastatin. –Adverse event rates in patients using rosuvastatin 40 mg were similar to rates in patients using atorvastatin 80 mg in short-term trials. 10

Key Question 1a: Statins New evidence for Update 5 20 new head-to-head trials of LDL cholesterol lowering drugs –3 double blind: atorvastatin vs. pravastatin; atorvastatin vs. simvastatin; atorvastatin vs. rosuvastatin –1 single blind –16 open label; 15 involving rosuvastatin 2 meta-analysis of head to head trials –Rogers 2007: atorvastatin vs. simvastatin –Wlodarczyk 2008: rosuvastatin vs. atorvastatin 11

Key Question 1a: Statins New evidence for Update 5 Percent LDL cholesterol lowering consistent with existing evidence Rosuvastatin 40 mg has a greater LDL cholesterol lowering effect than atorvastatin (53.6% to 58.8% vs. 46.3% to 55.4%) 12

Key Question 1a: FDCP New evidence for Update 5 13 active-controlled trials of LDL cholesterol lowering drugs: –10 active-control trials of ezetimibe/simvastatin: o 8 double blind, 8 comparing to statin, 1 with placebo and fenofibrate arms, 1 comparing to niacin o 2 open label comparing to doubling statin dose –2 double blind active-control trials of niacin extended- release/simvastatin compared to simvastatin 20mg –1 open label trial of niacin extended-release/lovastatin compared to atorvastatin 40mg and simvastatin 40mg 13

Key Question 1a: FDCP New evidence for Update 5 Ezetimibe-simvastatin 10/80 mg daily gives a 58.6% to 61.0% reduction in LDL cholesterol LDL cholesterol lowering of niacin extended- release fixed-dose combination products consistent with the statin effect alone 14

Key Question 1b How do statins and fixed-dose combination products containing a statin and another lipid lowering drug compare in their ability to reduce LDL cholesterol? Is there a difference in the ability of a statin or fixed-dose combination product containing a statin and another lipid lowering drug to achieve National Cholesterol Education Panel goals? 15

Key Question 1b: New evidence for Update 5 12 new statin trials and 8 fixed-dose combination trials Rosuvastatin 40mg may be more effective in meeting NCEP ATPIII LDL cholesterol goal of <100 at 24 weeks based on 1 new open-label trial comparing rosuvastatin 40 mg vs. atorvastatin 80 mg (83.6% vs. 74.6%) Higher doses of ezetimibe-simvastatin 10/40mg or 10/80 mg are more likely to meet NCEP ATP III LDL cholesterol goals when LDL cholesterol reductions >55% are needed Niacin-extended release fixed-dose combination products do not provide greater achievement in NCEP ATP III LDL cholesterol goals than the statin monotherapy 16

Key Question 2a How do statins and fixed-dose combination products containing a statin and another lipid lowering drug compare in their ability to raise HDL cholesterol? Are there doses for each statin or fixed-dose combination product containing a statin and another lipid lowering drug that produce similar percent increase in HDL cholesterol between statins? 17

Key Question 2a: Previous conclusions When statins are provided in doses that reduce LDL cholesterol by equivalent amounts, a similar percent increase in HDL cholesterol can be achieved There is conflicting evidence about simvastatin vs. atorvastatin, with some studies finding no difference and others finding simvastatin superior Some studies found greater increases in HDL cholesterol with rosuvastatin compared with atorvastatin, while other studies found no difference 18

Key Question 2a: New evidence for Update 5 2 open label head-to-head trials of rosuvastatin 40 mg compared to atorvastatin 80 mg (STELLAR, ECLIPSE) showed greater increase in HDL cholesterol with rosuvastatin 40 mg (8.4% to 10% vs. 1.8% to 3%). 4 open label head-to-head trials comparing low to moderate dose rosuvastatin to low to moderate dose simvastatin +/- atorvastatin showed greater increase of HDL cholesterol with rosuvastatin over atorvastatin but mixed results with simvastatin 19

Key Question 2a: New evidence for Update 5 Ezetimibe-simvastatin has an equivalent effect on increasing HDL cholesterol as equipotent simvastatin monotherapy Extended-release niacin/lovastatin and ER niacin/simvastatin fixed-dose combination products are more effective in raising HDL cholesterol than equipotent statin monotherapy NER/S 2000/20 mg 24.9%, NER/L 2000/40 mg 32%) Niacin extended-release fixed-dose combination products have increased adverse events and greater discontinuation of drug due to adverse event (15.6% vs. 5.3% simvastatin) 20

Key Question 2b How do statins and fixed-dose combination products containing a statin and another lipid lowering drug compare in their ability to raise HDL cholesterol? Is there a difference in the ability of a statin or fixed-dose combination product containing a statin and another lipid lowering drug to achieve National Cholesterol Education Panel goals? 21

Key Question 2b: New Evidence for Update 5 19 new head-to-head trials and 1 new meta-analysis of statins addressing HDL effect 13 fixed-dose combination product active-control trials Ezetimibe-simvastatin equivalent to simvastatin alone One 24-week double blind trial of niacin ER/simvastatin fixed-dose combination product showed greater ability to meet HDL cholesterol NCEP goals (70% vs. 45%) but greater discontinuation of drug due to adverse event (15.6% vs. 5.3% simvastatin) 22

Key Question 3 How do statins and fixed-dose combination products containing a statin and another lipid lowering drug compare in their ability to reduce the risk of nonfatal myocardial infarction, coronary heart disease (angina), coronary heart disease mortality, all-cause mortality, stroke, hospitalization for unstable angina, or need for revascularization (coronary artery bypass graft, angioplasty, or stenting)? 23

24 Key Question 3: Primary prevention Still no direct, head-to-head evidence No studies of fixed-dose combination products No changes to previous conclusions this update

25 Key Question 3: New evidence in primary prevention JUPITER: –Rosuvastatin 20 mg vs. placebo –Multicenter, N=17 802, normal lipid levels, elevated C-reactive protein; scheduled 5-year follow-up, stopped early (median 1.9 years) for benefit –Lower risk of first major CV event by 44% (hazard ratio, 0.56; 95% CI, 0.46 to 0.69) –Absolute risk 1.6% vs. 2.8% (number needed to treat, 83) –Number needed to treat for all-cause mortality, 167 –Extrapolating results beyond 1.9 years not recommended MEGA: -Lower-dose pravastatin mg plus diet vs. diet alone; Japanese adults -33% relative reduction in incidence of coronary events at 5.3 years followup -All cause mortality lower but not significant (hazard ratio, 0.72 (95% CI, 0.51 to 1.01)

26 Key Question 3: New evidence in patients with diabetes 8 trials (2 new this update) evaluated long-term effectiveness of atorvastatin, simvastatin, and fluvastatin ASPEN: atorvastatin 10 mg, 4 years: no reduction in risk of primary outcome, in contrast to 3 other studies showing reduced risk Xu: fluvastatin 80 mg, post PCI, reduced risk of primary outcome, no reduction in all-cause mortality

27 Key Question 3: New evidence in secondary prevention SPARCL: N=4731, recent stroke or TIA Atorvastatin 80 mg vs. placebo; 4.9 years follow-up Reduced risk of stroke (hazard ratio, 0.84; 95% CI, 0.71 to 0.99) 1.9% absolute risk reduction (number needed to treat, 53) Patients with ischemic or unclassified type of stroke benefited the most; those with hemorrhagic type more likely to have a harmful event (hazard ratio, 1.66; 95% CI, 1.08 to 2.55) Reduced risk of major coronary events and need for revascularization No decrease in CVD death or all-cause mortality

Key Question 4 Are there differences in effectiveness of statins and fixed-dose combination products containing a statin and another lipid lowering drug in different demographic groups or in patients with comorbid conditions (e.g., diabetes, obesity)? 28

Key Question 4 Four meta-analyses (2 new this update) suggest statins are equally efficacious in men, women, and elderly patients 3 new studies confirmed earlier evidence that rates of adverse events do not differ among men, women, and elderly patients at maximum doses of simvastatin and lovastatin 29

Key Question 5 Are there differences in the harms of statins or fixed-dose combination products containing a statin and another lipid lowering drug when used in the general population of children or adults? 30

Key Question 5 Previous update: –There is insufficient evidence to determine which statin or statins are safer with regard to muscle toxicity or elevated liver enzymes. New evidence: –Niacin extended-release fixed-dose combination products have increased adverse events, primarily flushing, with greater discontinuation of drug due to adverse event 31

Key Question 6 Are there differences in the harms of statins or fixed-dose combination products containing a statin and another lipid lowering drug when used in special populations or with other medications (drug-drug interactions)? 32

33 Key Question 6: Previous conclusions Studies in patients with diabetes did not have higher rates of adverse events Potential for interactions with CYP 3A4 inhibitors (atorvastatin, lovastatin, and simvastatin) Potential for interaction with CYP 2C9 inhibitors (fluvastatin) Statin-fibrate combination increases risk of musculoskeletal-related adverse events compared with monotherapy No changes to previous conclusions this update

34 Summary and conclusions: Adults Key Questions 1 and 2 (lipids) –High dose rosuvastatin appears to have greater LDL cholesterol lowering and HDL raising effect than high dose atorvastatin –Ezetimibe-simvastatin fixed-dose combination product has greater LDL cholesterol lowering than statin monotherapy –Niacin ER fixed-dose combination products have greater HDL raising than statin monotherapy Key Question 3 (health outcomes) –Still no head-to-head evidence for equipotent doses of statins –No evidence for fixed-dose combination products

35 Key Question 4 (comparative effectiveness in subgroups) −No changes to conclusions this update −Little comparative evidence Key Question 5 (harms in the general population −Niacin ER fixed-dose combination products have greater adverse events, primarily flushing Key Question 6 (harms in special populations) −No changes to conclusions this update Summary and conclusions: Adults

36 Results: Children

37 Key Question 1 How do statins and fixed-dose combination products containing a statin and another lipid lowering drug compare in their ability to reduce LDL cholesterol among children?

38 Trials of statins in children have been conducted primarily in children with heterozygous or homozygous familial hypercholesterolemia, or other familial dyslipidemias 8 trials of various statins showed improvement in LDL cholesterol compared with placebo Key Question 1

39 In meta-analysis, statins reduced LDL cholesterol in children taking a statin by 32% (95% CI, 37 to 26) One trial compared ezetimibe/simvastatin to simvastatin alone and demonstrated a 54% reduction in LDL cholesterol for combination compared to 38% reduction for simvastatin alone Key Question 1

40 How do statins and fixed-dose combination products containing a statin and another lipid lowering drug compare in their ability to raise HDL cholesterol among children? Key Question 2

41 Key Question 2 Statins decreased HDL cholesterol in one study of atorvastatin and did not change HDL cholesterol in 5 other trials of statins including rosuvastatin, simvastatin, lovastatin, and pravastatin Overall, the pooled result indicated that statins increased HDL cholesterol by 3% (95% CI, 0.6 to 5.6)

42 Key Question 3 How do statins and fixed-dose combination products containing a statin and another lipid lowering drug compare in their ability to reduce the risk of major coronary heart disease outcomes (morbidity and mortality) among children?

43 Key Question 3 Studies of statins in children have not been conducted with long enough follow-up to assess for outcomes related to cardiovascular mortality and morbidity

44 Key Question 4 Are there differences in effectiveness of statins and fixed-dose combination products containing a statin and another lipid lowering drug among children different demographic groups or in patients with comorbid conditions (e.g. diabetes, obesity)?

45 Key Question 4 No trials have evaluated statins in children with diabetes or obesity One study demonstrated 21% reduction in LDL cholesterol with simvastatin in children with neurofibromatosis 1

46 Are there differences in the harms of statins or fixed-dose combination products containing a statin and another lipid lowering drug when used in the general population of children? Key Question 5

47 Key Question 5 Adverse events were variably reported; methods of detection and assessment of adverse events were often not specified Multiple studies reported no significant elevations in either CK and AST/ALT Elevations in AST/ALT occurred but were either lower than 3 times the upper limit of normal, or resolved with interruption/discontinuation of medication

48 Key Question 6 Are there differences in the harms of statins or fixed-dose combination products containing a statin and another lipid lowering drug when used in special populations or with other medications (drug-drug interactions)?

49 Key Question 6 One study of fluvastatin in children with minimal change glomerulonephritis demonstrated decrease in total cholesterol and reported no side effects.

Oregon Evidence-based Practice Center Mark Helfand, MD, MPH, EPC Director Drug Effectiveness Review Project Marian McDonagh, PharmD, Principal Investigator Oregon Health & Science University