Should we treat everybody over 60 years with a statin? Comprehensive primary prevention in practice

Pathogenesis of atherosclerosis A decades-long disease course Inflammation Selectins ICAM IL M-CSF CRP Endothelial stress BP Smoking VLDL LDL HDL 1 2 3 4 5 6 7 Age decade Libby (2001) Circulation 104:365

Comprehensive primary prevention in practice Assessing the risks and benefits of age-determined intervention for all. The evidence base for primary prevention. Optimising primary prevention strategies.

Impact of age on CHD risk Ridker et al (2016) Europ Heart J. June online

EAS/ ESC Guidelines 2011 Concept of age linked intervention Catapano et al (2011) Atherosclerosis 217:3-46

Perceptions of risk

Response to UK NICE widening of statin use in primary prevention - 2014 Statins: Millions more to get drugs in controversial plans Comments (235) 2 May 2014 By James Gallagher Health editor, BBC News website Doctors have been told to offer cholesterol-lowering statins to millions more people in a massive and controversial expansion in prescribing. Four in 10 adults in England, Wales and Northern Ireland are now eligible for statins, even though many are at low risk of a heart attack or stroke. May 15 th 2016 But cardiologist Dr Aseem Malhotra said: 'I have no doubt millions of people taking statins in the UK will not benefit but are being put at risk of unnecessary harm.' Concerns about statins were raised earlier this year by the Queen's former doctor, Sir Richard Thompson.

Assessing statin intolerance GAUSS -3 experience All subjects in Phase A identified as statin intolerant (80% to 3 agents) Nissen et al. JAMA. 2016;315:1580-1590

Disease trajectories in CHD prevention Asymptomatic phase Plaque rupture Postponement of coronary event Statin Rx Statin Plus Rx Clinical event horizon Total modifiable risk Residual modifiable risk Non-modifiable risk age, sex, genetics Fatty streak Unstable lesion Stable lesion 30 40 50 60 70 80 Age (years) Packard CJ, Weintraub WS, Laufs U. Vascul Pharmacol. (2015); 71:37-9

Placebo MI rate per 100 subjects per 5 years Assessing benefit of lipid lowering Placebo - Statin outcome trials Continuum of risk 53.7 22.6 12.9 8.44 7.9 2.8 End stage Secondary prevention Primary prevention CORONA GISSI-HF (rosuvastatin) 4S (simvastatin) HPS (simvastatin) CARE (pravastatin) LIPID (pravastatin) PROSPER (pravastatin) WOSCOPS (pravastatin) AFCAPS/TexCAPS (lovastatin) JUPITER (rosuvastatin) Heart failure High-risk CHD patients (high cholesterol) Majority of CHD patients (broad range of cholesterol levels) Patients at high risk of CHD (high cholesterol) Patients at low risk of CHD (low HDL-C)

Relative risk reduction Lessons from completed LDL lowering trials Risk reduction is related to LDL decrease Data from trials of:- Statin vs placebo More vs less intense statin therapy. Combination therapy with ezetimibe LDLc reduction (mmol/l) Regression line reveals:- 1.0 mmol/l fall in LDLc translates into a 22% decrease in risk CTTC Lancet (2005) 367;1267-78 Cannon et al NEJM (2015) 367; June

Modelling impact of cholesterol lowering therapy at age 60 y. Rx Atorvastatin 20mg daily Average LDL = 130 mg/dl (3.4 mmol/l) LDL decrease = 45% (1.53 mmol/l) Relative risk reduction = 33% Age 60 years 10 year CHD event rate = 18% (SCOREx3) Events prevented per 1000 = 60 NNT 10 years = 16 Based on Prospective Studies Collaboration Lancet (20107) 370:1829-39

Percentage with event Assessing long term (lifetime) benefits of LDL lowering in WOSCOPS CHD mortality All-cause mortality Heart failure 12 10 8 6 4 2 0 Over entire period 27% risk reduction P<0.001 Placebo Original trial Pravastatin 0 2 4 6 8 10 12 14 16 18 20 22 Years since randomisation 55 65 75 y Average age of cohort 45 40 35 30 25 20 15 10 5 0 Over entire period 13% risk reduction P<0.001 Placebo Pravastatin 0 2 4 6 8 10 12 14 16 18 20 22 Years since randomisation 5 4 3 2 1 0 Over 20 years HR 0.69, P=0.0068 0 2 4 6 8 10 12 14 16 18 20 22 Years since randomisation Ford I et al. Circulation. (2016); 133:1073-80

Long term safety in statin studies WOSCOPS 20 year experience Placebo, number (%) with event Pravastatin, number (%) with event Adjusted Hazard Ratio Endpoint Total n=3293 Total n=3302 (95% Confidence Interval) P- Value All cancers 816 ( 24.8%) 809 ( 24.5%) 0.96 ( 0.87, 1.06) 0.41 Colorectal cancer 140 ( 4.25%) 127 ( 3.85%) 0.87 ( 0.68, 1.10) 0.25 Lung cancer 202 ( 6.13%) 187 ( 5.66%) 0.89 ( 0.73, 1.08) 0.24 Prostate cancer 170 ( 5.16%) 186 ( 5.63%) 1.05 ( 0.85, 1.29) 0.65 Upper GI cancer 77 ( 2.34%) 87 ( 2.63%) 1.09 ( 0.80, 1.48) 0.60 Urinary tract cancer 97 ( 2.95%) 99 ( 3.00%) 0.99 ( 0.75, 1.31) 0.93 Other cancer 160 ( 4.86%) 157 ( 4.75%) 0.95 ( 0.76, 1.18) 0.62 All non-cvd deaths 757 ( 23.0%) 731 ( 22.1%) 0.92 ( 0.83, 1.02) 0.12

WOSCOPS - in trial and post trial event rates Placebo, number (%) with event Total n = 3293 In trial event rates (1989-1995) In trial Pravastatin, number (%) with event Adjusted Hazard Ratio (95% CI) Endpoint Total n = 3302 P-value* Fatal or nonfatal MI 190 ( 5.77%) 115 ( 3.48%) 0.59 ( 0.47, 0.74), <0.0001 Placebo, number (%) with event Post trial Pravastatin, number (%) with event Adjusted Hazard Ratio (95% CI) Total n = 3023 Total n = 3118 P-value* 427 ( 14.13%) 372 ( 11.93%) 0.82 ( 0.71, 0.94), 0.0054 CHD related death or nonfatal MI 198 ( 6.01%) 119 ( 3.60%) 0.58 ( 0.47, 0.73), <0.0001 480 ( 15.88%) 418 ( 13.41%) 0.82 ( 0.72, 0.93), 0.0028 CHD related death or hospitalisation 273 ( 8.29%) 177 ( 5.36%) 0.58 ( 0.47, 0.72), <0.0001 823 ( 27.92%) 739 ( 24.12%) 0.79 ( 0.70, 0.89), 0.0002 CV related death or hospitalisation 415 ( 12.60%) 329 ( 9.96%) 0.62 ( 0.52, 0.73), <0.0001 1301 ( 46.05%) 1215 ( 41.51%) 0.81 ( 0.73, 0.90), <0.0001 Fatal or nonfatal stroke 40 ( 1.21%) 29 ( 0.88%) 0.56 ( 0.31, 1.03), 0.0608 332 ( 10.61%) 329 ( 10.36%) 1.00 ( 0.82, 1.22), 0.9856 Ford I et al. Circulation. (2016); 133:1073-80

Cumulative CHD burden 20 years in 72,000 WOSCOPS screened men Group (n) Mean cholesterol [LDL] (mmol/l) Observed CHD hospitalisation events per 100 subjects over 20 years Observed CHD hospital days per 100 subjects over 20 years No CHD history primary prevention Adjusted hazard ratio a (95% CI) P4 (10767) 7.05 [5.0] 52.7 349.6 2.2(2.0-2.5) NNT = 3-4 P3 (22288) 5.98 [4.0] 42.4 288.3 1.8(1.7-2.0) P2 (18952) 5.06 [3.1] 33.6 232.2 1.5(1.3-1.6) P1 (7414) 4.00 [2.1] 23.0 167.2 1.0(referent) With CHD history secondary prevention S4 (2210) 7.10 [5.1] 152.9 1089.9 1.9(1.6-2.4) NNT = 1-2 S3 (3105) 6.02 [4.0] 140.6 931.7 1.8(1.5-2.2) S2 (2037) 5.08 [3.1] 113.8 773.4 1.5(1.2-1.8) S1 (741) 3.98 [2.0] 77.3 526.1 1.0(referent)

Pathogenesis of atherosclerosis Asymptomatic early lesions Clinically significant pathology Raised fatty streaks Complex/ ulcerated plaques

Percentage with event Intervention versus prevention Non-fatal MI/ CHD death 24 Gain in event free years 21 18 5 years 15 12 9 6 placebo pravastatin 3 0 0 2 4 6 8 10 12 14 16 18 20 22 Years since randomisation

Costs decrease Costs increase Benefits of primary prevention Clinical benefit is evident in low risk populations Statin therapy is cost-saving over long term in primary prevention WOSCOPS Cumulative costs per 1000 statin treated subjects (CVD Hospitalisation; drug, monitoring) 200k 0 200k 400k 600k 800k 1000k 1200k CTTC Lancet May 2012 (on-line) McConnachie, Ford, Packard et al. 2013 (WOSCOPS,unpublished)

Troponin as an index of cardiomyocyte damage or stress Troponin T, I, C complex with actino-myosin in cardiac muscle

Troponin I and CVD risk in a primary prevention population - WOSCOPS Troponin I measured at baseline (pre treatment) Q4 vs Q1 HR 2.27 Ford et al JACC (2016) 68:2719-28

Dynamic measures of atherosclerosis progression Troponin I measured at baseline and Year 1 on treatment. HR 5 to 6 for primary endpoint for Q4 vs Q1. Why is change in trop I a stronger predictor of risk? TnI increased over Year 1 TnI decreased over Year 1 Ford et al JACC (2016) 68:2719-28

LDL lowering in primary prevention Targeting based on gene score Khera et al N.Eng. J Med. (2016) Nov 13 Mega et al Lancet (2015) 385:2264-71

An age-based plan for comprehensive primary CVD prevention Selective screening. Family history of CHD. Genetic testing Risk factor + Biomarker (TnI, BNP) assessment. Bi-annual check. Offer statin to all. Compliance. Side effects.

Summary Age is the single most important determinant of CHD risk. An argument can be made for universal statin therapy initiation at a age linked to high risk (7.5% to 10%/ 10 years). Benefits clear from trial evidence base long term safety and efficacy. Risk appreciation and adherence are challenges. Strategy is clinically effective but socially acceptance is yet to be proven.

End 45 mins

LDL and atherosclerosis A coalescence of evidence Genomics Genetics Epidemiology Pathology Intervention trials

Pharmacological CHD prevention in populations Prescribe statins to those at risk at 40, study says (July 2009)

EAS/ ESC Guidelines 2011 Impact of age on CHD risk Catapano et al (2011) Atherosclerosis 217:3-46

Role of LDL in atherogenesis Cellular remodelling Release of bio-active lipids Extracellular pathology Lipoprotein retention Deposits early microand later macrocholesterol crystals. Innate immunity Initiates localised inflammation. Macrophage recruitment. Intracellular cholesterol accumulation. Necrosis Auto-immunity (apob) Adaptive immunity See Libby, Ridker, Hansson. Nature (2011); 473:317-325

Event rate (%) Association of LDL cholesterol with CHD risk in statin trials 30 4S - Placebo 25 20 Rx - Statin therapy PRA pravastatin ATV - atorvastatin 4S - Rx Secondary Prevention 15 10 5 0 20 (0.5) LIPID - Rx CARE - Placebo CARE - Rx HPS - Rx TNT ATV10 HPS - Placebo TNT ATV80 PROVE-IT - PRA WOSCOPS Placebo PROVE-IT ATV AFCAPS - Placebo IMPROVE-IT sim IMPROVE-IT e/s 6 AFCAPS - Rx WOSCOPS - Rx JUPITER -RSV JUPITER - Placebo ASCOT - Placebo ASCOT - Rx 40 (1.0) 60 (1.6) 80 (2.1) 100 (2.6) 120 (3.1) LDL-C achieved mg/dl (mmol/l) 140 (3.6) LIPID - Placebo 160 (4.1) Primary Prevention 180 (4.7) 200 (5.2) Adapted from Rosensen RS. Exp Opin Emerg Drugs 2004;9(2):269-279 LaRosa JC et al. N Engl J Med 2005;352:e-version

Relative risk reduction Exploring LDL lowering strategies using inherited variation LDL lowering due to variation in:- Cholesterol absorption (ezetimibe) Cholesterol production (statin) LDL receptor number (PCSK9 inhibitor) LDLc reduction (mmol/l) CHD risk reduction per unit LDL decrease is independent of intervention target. Ference et al JACC (2015) March on-line

Inherited vs pharmacologically based LDL lowering Earlier is better Ference et al JACC (2015) March (0n-line)

Event (%) WOSCOPS: Early Event Reductions With Pravastatin 12 10 8 6 4 2 0 1 Myocardial Infarction Placebo (n = 3293) Pravastatin (n = 3302) 2 3 4 5 6 Years in Study 31% risk reduction P < 0.001 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 Cardiovascular Mortality Placebo (n = 3293) Pravastatin (n = 3302) 0 1 2 3 4 5 6 Years in Study 32% risk reduction P = 0.033 6 5 4 3 2 1 0 1 Total Mortality Placebo (n = 3293) Pravastatin (n = 3302) 2 3 4 5 6 Years in Study 22% risk reduction P = 0.051 3.0 2.5 2.0 1.5 1.0 0.5 0.0 PTCA/CABG Placebo (n = 3293) Pravastatin (n = 3302) 0 1 2 3 4 5 Years in Study 37% risk reduction P = 0.009 Shepherd et al. N Engl J Med. 1995;333:1301-1307.

% Long term safety of statins WOSCOPS: 15 year follow up CHD-related death or nonfatal MI Risk reduction in statin group 40% during trial (P<0.001) 18% post-trial (P=0.02) 27% overall follow-up (P<0.001) 70 60 50 40 30 20 10 0 0 1 2 3 4 5 Years post-trial Placebo Pravastatin K-M curves according to the originally assigned study group Ford I et al. N Engl J Med 2007;357:1477-1486

Options in LDL lowering High dose statin vs. combination therapy 21 st March 2014 AE profile High vs moderate dose statin Event Odds ratio(ci) Any AE 1.44(1.33-1.55) LFT abnormalities 4.48(3.27-6.16) CK>10 9.97(1.28-77.9) Rhabdomyolysis 1.66(0.60-4.57) Silva et al (2007) Clin. Therap. 29:253-260

Percentage with event Additional/ long term benefits of LDL lowering in heart failure Effect of statin therapy on non-fatal/ fatal heart failure in 14 LDL lowering trials WOSCOPS Heart failure hospitalisations 5 4 Over 20 years HR 0.69, P=0.0068 3 2 1 Overall (I 2 =0.0%, P=0.757) 0.90 (0.84, 0.97) 0 0 2 4 6 8 10 12 14 16 18 20 22 Years since randomisation 0.2 1.0 5.0 Relative risk Preiss et al EHJ (2015) March on-line

Potential mechanism of a legacy effect in LDL lowering trials (HPS) (WOSCOPS) Packard, Ford Curr Opin Lipidol (2015) 26;572-9

Troponin I is decreased by statins LDL lowering with pravastatin moved subjects to lower quintiles of change in trop I.? LDL lowering induced net repair.? Change in trop I as biomarker of atherosclerosis progression. Ford et al JACC (2016) 68:2719-28

Legacy benefit in statin trials - ASCOT Lipid changes Primary endpoint Sever et al Europ Heart J(2008) 29;499-508

Damage and repair in the natural history of atherosclerosis Troponin I in asymptomatic subjects is a measure of micro-damage possibly ruprure and repair of atheromatous lesions Change in trop I is a stronger predictor of because it identifies repairers and nonrepairers. Rupture Trop I Small lesion Microthrombus Repair Healed plaque Trop I

Legacy benefit in statin trials - HPS HPS Collaborative Group Lancet (2011) 378;2013-2020

Emerging Risk Factors Collaboration Plasma lipids and incident CHD 302,430 subjects; 2.79 million pt-yrs; 8857 MIs; 3928 strokes; 68 studies Danesh et al (2009) JAMA 302:1993-2000

CHD (fatal +non-fatal) event rate per 5 years - Secondary prevention trials CHD (fatal +non-fatal) event rate per 5 years - Primary prevention trials Achieved LDL in primary and secondary prevention trials 9,0 25,0 4S-p WOSCOPS-p 8,0 20,0 15,0 10,0 5,0 ASCOT-p WOSCOPS-a LIPID-p PROSPER-a IDEAL-s PROSPER-p CARE-p CARE-a HPS-p IDEAL-a AFCAPS-a LIPID-a ASCOT-a HPS-a TNT-10 JUPITER-p TNT-80 JUPITER-a 4S-a AFCAPS-p 7,0 6,0 5,0 4,0 3,0 2,0 1,0 0,0 0,0 0 50 100 150 200 250 Achieved LDL cholesterol (mg/dl)

Clinical Benefit of Lowering LDL-C Earlier Baseline LDL-C mg/dl (mmol/l) LDL-C after 50% reduction RR for CHD* Relative Reduction in Lifetime Risk 190 (4.9) 95 (2.5) 0.15 85% 160 (4.1) 80 (2.1) 0.20 80% 140 (3.6) 70 (1.8) 0.25 75% 120 (3.1) 60 (1.6) 0.30 70% 100 (2.6) 50 (1.3) 0.37 63% 80 (2.1) 40 (1.0) 0.45 55% * Based on 54% RRR per mmol/l lifetime exposure to lower LDL-C. Calculated as RR = 0.46 (Δ in mmol/l) ACC/AHA Guidelines recommend treating threshold of 190 mg/dl However, similar very large lifetime risk reduction for persons with LDL-C well below 190 mg/dl Ference, BA et al. J Am Coll Cardiol 2012;60:2631-9. Ference, BA et al. J Am Coll Cardiol 2015;651552 61.

Long term benefit in statin trials PROSPER long-term followup (Scottish cohort) Coronary death or hospitalisation for MI Endpoint Placebo (n) In trial Post trial Pravastatin (n) Placebo (n) Pravastatin (n) Original trial Placebo CHD+stroke death or admission 135 113 256 256 Risk ratio 0.80 (0.62-1.03) 0.96 (0.81-1.15) Pravastatin 75 85 y Average age of cohort CHD death or CHD admission 143 111 197 186 Risk ratio 0.74 (0.58-0.95) Lloyd et al, PLoS ONE (2013) 8; e72642 0.86 (0.70-1.05) Does the strength of the legacy benefit depend on the age treatment is started?

Disease trajectories in CHD prevention Asymptomatic phase Plaque rupture Postponement of coronary event Gain in Event Free Years Statin Rx Statin Plus Rx Clinical event horizon Primordial prevention of atherogenesis Total modifiable risk Residual modifiable risk Non-modifiable risk age, sex, genetics Fatty streak Unstable lesion Stable lesion 30 40 50 60 70 80 Age (years)