Craig Williams, Pharm.D., FNLAAssociate Professor; OSU School of Pharmacy and OHSU School of Medicine

williacr@ohsu.edu

2011 Annual Diabetes Practice Update

Session:

Update on Cardiovascular Disease & Diabetes:

Prevention of CVD in Patients with Diabetes

September 30th, 2011

Craig Williams, Pharm.D., FNLAAssociate Professor; OSU School of Pharmacy and OHSU School of Medicine

williacr@ohsu.edu

Presenter Disclosure InformationIn compliance with the accrediting board policies, theAmerican Diabetes Association requires the following

disclosure to the participants:

Name of Presenter: Craig Williams, Pharm.D., FNLA

Research Support: Merck & Co., Inc.

Speakers Bureau: Merck & Co., Inc.

VI. Prevention and management of complications A. Cardiovascular disease 1. HTN control 2. Dyslipidemia management 3. Anti-platelet 4. Smoking cessastion

5. Hyperglycemia?

NEW ENGLAND JOURNAL OF MEDICINE March 4, 2010

Glycated Hemoglobin, Diabetes, and Cardiovascular Risk in Non-diabetic AdultsElizabeth Selvin, Michael Steffes, Hong Zhu, Kunihiro Matusushita, et al.

There is a clear epidemiologic association between glycemic control and CVD

Data from 11,092 black and white subjects in the ARIC trial (Atherosclerosis Risk in Communities)

Median follow approximately 14 years.

ADA position on glycemia and macrovascular disease in 2010 Standards of Care guideline

ADA Standards of Care. Diabetes Care 2010;33:S11-62

Despite clear epidemiology, controversy continues regarding the role of glucose lowering to prevent coronary events

Three large trials of glycemic control published in 2008 failed to find CVD benefit

Sklyer JS, et al. Intentive glycemic control and the prevention of cardiovascular events. A position statement of the ADA/ACC/AHA. Diabetes Care 2009;32:187-92.

So hyperglycemia doesn’t matter to the heart?

Non-fatal MI significantly reduced 24% (p=0.001)

Retnakaran R, Zinman B. Lancet 2008;371:1790-99.

Hyperglycemia is toxic at several steps in the atherosclerosis process

Failure to find benefit may have related to the A1C levels tested:

6.4% vs. 7.5% 6.3% vs. 7.0% 6.9% vs. 8.5%

Sklyer JS, et al. Intentive glycemic control and the prevention of cardiovascular events. A position statement of the ADA/ACC/AHA. Diabetes Care 2009;32:187-92.

So hyperglycemia matters to the heart but intense control (A1C < 7%) provides little additional benefit over moderate control (A1C 7-8%)

2011 ADA guideline appropriately discusses microvascular benefits of A1C < 7% while acknowledging lack of proven macrovascular benefits at the A1C values that were studied.

VI. Prevention and management of complications A. Cardiovascular disease 1. HTN control 2. Dyslipidemia management 3. Anti-platelet 4. Smoking cessastion

2. Dyslipidemia/ lipid management

ADA Standards of Care; Diabetes Care, January 2011

“Alternative goals…..”

Three points critical to understanding the evidence base of the ADA guidelines for lipid management:

1.The etiologic role of lipoproteins in atherosclerosis

2.The etiology of dyslipidemia as seen in patients with diabetes

3.The clinical outcomes literature in patients with diabetes

1. Atherosclerosis is a lipoprotein driven processBasic Science for Clinicians

Subendothelial Lipoprotein Retention as the Initiating Process in AtherosclerosisUpdate and Therapeutic Implications

Ira Tabas, MD, PhD; Kevin Jon Williams, MD; Jan Borén, MD, PhD

Circulation, October 16th, 2007

Lipoproteins share structural homology

Chylomicrons, VLDL, IDL, LDL, HDL all share a basic biochemistry

Liver

VLDL

TG

IDL

LDL

Lipase enzymes

Lipase enzymes

LDLc

Type (%)

Appearance of serum

Elevated particles

Associated clinical disorders TC TG

I (~1%) Creamy top layer Chylomicrons,VLDL

Lipoprotein lipase deficiency, apolipoprotein C-II deficiency

+ +++

IIa (10%) Clear LDL Familial hypercholesterolemia, polygenic hypercholesterolemia, nephrosis, hypothyroidism, familial combined hyperlipidemia

++ ↔

IIb (40%) Clear LDL, VLDL Familial combined hyperlipidemia ++ +

III (~1%) Turbid IDL Dysbetalipoproteinemia + +

IV (45%) Turbid VLDL Familial hypertriglyceridemia, familial combined hyperlipidemia, sporadic hypertriglyceridemia, diabetes

+ ++

V (5%) Creamy top, turbid bottom

Chylomicrons, VLDL (remnants)

Diabetes + ++

Fredrickson Classification of DyslipidemiaWe look at this

Artery wall sees these

LDLLDL

LDLLDLEndotheliumEndothelium

Vessel LumenVessel LumenMonocyteMonocyte

MacrophageMacrophage

MCP-1MCP-1

AdhesionAdhesionMoleculesMolecules

Steinberg D et al. N Engl J Med 1989;320:915-924.

The primary atherogenic lipoprotein is LDLipoprotein is LDLlipoproteins of > 70 nm have limited transcytosis past the endotheliumlipoproteins of > 70 nm have limited transcytosis past the endothelium

The primary atherogenic lipoprotein is LDLipoprotein is LDLlipoproteins of > 70 nm have limited transcytosis past the endotheliumlipoproteins of > 70 nm have limited transcytosis past the endothelium

Foam CellFoam Cell

Modified LDL Modified LDL Taken up by Taken up by MacrophageMacrophage

IntimaIntima

Nascentchylomicron

Nascet VLDL

ΧΧ

Artery wall

Am Heart J 2008;156:112-119

2. Dyslipidemia vs. hyperlipdemia: Prevalence in NHANES 2008 data: High TG or low HDLc more

common than high LDLc

Prevalence of Dyslipidemia is high in Type 2 Diabetes

Jacobs MJ, et al. Diabetes Res Clin Pract. 2005;70:263-269.

Control of Lipids Patients With Diabetes, %

Patients Without Diabetes, %

P Value

LDL-C > 100 mg/dL 74.7 75.7 NS

HDL-C < 40 mg/dL (men)< 50 mg/dL (women)

63.7 40.0 < .001

Triglycerides> 150 mg/dL 61.6 25.5 < .001

N = 498 adults (projected to 13.4 million) aged > or = 18 years with diabetes representative of the US population and surveyed within the cross-sectional National Health and Nutrition Examination Survey

1999-2000.

Hepatic lipase

Fat Cells Liver

Kidney

Insulin

CETPCETPCETPCETP

CE

VLDL HDL

Lipoprotein lipaseor hepatic lipase

Small, denseSmall, denseLDLLDLLDLLDL

TGApoA-I

TGTGCE

FFA

‘Dyslipidemia’ is a state of relative insulin resistance resulting in a conversion of adipose tissue to an exocrine state. Excessive production of free fatty acids (FFA) increases hepatic VLDL production

CE, cholesteryl esters; FFA, free fatty acids; TG, triglycerides.Ginsberg HN. J Clin Invest. 2000;106:453–458.

CETPCETPCETPCETP

↑ TG↑ ApoB↓ HDLc↔ LDLc

XInsulinInsulinresistanceresistance

Liver

IDLIDL

FFA

While LDLc is similar, particle burden is heavier

LDL particle count vs. cholesterol contentTo carry the same amount of cholesterol, a larger number of particles are needed if

they are smaller

apoB is a measure of number of atherogenic lipoproteins (essentially VLDL, IDL, LDL). Non-HDL is measure of cholesterol carried in these same particles

LDLc measures cholesterol carried in LDL and IDL

Small, dense: 25-30 nmLarge, buoyant: 30-35 nm

LDLc=115 mg/dl LDLc=115 mg/dl

Summary: Patients with diabetes have elevated TG and lower HDLc but also a greater number of LDL particles which confers greater risk at any measured LDLc value

3. What are the data for LDLc lowering?

ADA guidelines: Major statin trials or sub-studies in diabetic patients

Lancet 2004;364:685Diabetes Care 2006;29:1220Lancet 2003;361:2005Diabetes Care 2006;7:1478Diabetes Care 1997;20:614

*Num. needed to treat (NNT) for moderate-high risk DM to avoid one death or MI:

3-50

ADA Standards of Care; Diabetes Care, January 2011

Reduction in 10-year CVD events with statin therapy in patients with diabetes: Event reduction correlates with relative risk – more risk, more benefit

Endpoint: 10-year Fatal CHD/Non-fatal MI and LDL lowering Relative Risk reduction ARR LDL reduction

4S-DM 85.7 to 43.2% (50%) 42.5% 186 to 119 mg/dL (36%)

ASPEN 20 35.1 to 23.2% (34%) 11.9% 112 to 79 mg/dL (29%)

HPS-DM 20 43.8 to 36.3% (17%) 7.5% 123 to 84 mg/dL (31%)

CARE-DM 40.8 to 35.4% (13%) 5.4% 136 to 99 mg/dL (27%)

TNT-DM 26.3 to 21.6% (18%) 4.7% 99 to 77 mg/dL (22%)

HPS-DM 10 17.5 to 11.5% (34%) 6.0% 124 to 86 mg/dL (31%)

CARDS 11.5 to 7.5% (35%) 4% 118 to 71 mg/dL (40%)

ASCOT-DM 11.1 to 10.2% (8%) 0.9% 125 to 82 mg/dL (34%)

ASPEN 10 9.8 to 7.9% (19%) 1.9% 114 to 80 mg/dL (30%)

10: Primary prevention data 20: Secondary prevention

2○

1○

The differential benefit of LDLc lowering in patients with diabetes has been evident from the earliest statin trials and is more evidence that higher risk=greater benefit : 4S study: Major Coronary Events

1111 2222 3333 4444 5555 6666000050505050

80808080

90909090

100100100100

55%55%

0000

Diabetic – simvastatinDiabetic – placeboNondiabetic – simvastatinNondiabetic - placebo

Diabetic - simvastatin

Diabetic - placebop=0.002

Risk reduction

Coronary Death and non-fatal MI

Years since randomizationPyörälä K, et al. Pyörälä K, et al. Diabetes CareDiabetes Care. 1997;20:614–620. 1997;20:614–620

Per

cent

of

patie

nts

with

out m

ajor

CV

eve

nt

70707070

Within a given population, lower goals do further reduce CVD events: Risk Curve ConceptHigher risk patients have more to gain from aggressive therapy

Robinson JG, Stone NJ. Am J Cardiol. 2006;98:1405-1408

0

Car

diov

ascu

lar

Eve

nt R

ate

(%)

0 20 40 60 80 100 120 140 160 180 200

LDL (mg/dL)

No CVD - No diabetes

Diabetes - No CVD

CHD - NoMS or IFG

CHD + MS or IFG

CHD + Diabetes

80

70

60

50

40

30

20

10

What aggressive LDL lowering does: reduces atheroma volume in arterial wall providing plaque ‘stabilization’

Brown et al. Arter Thromb Vasc Biol 2001;21:1623

Treated: LDLc of 84 mg/dL (47% reduction)Untreated: LDLc of 163 mg/dL with statin+resin

N 4,444 9,014 4,159 20,536 6,595 6,605∆LDL -36% -25% -28% -29% -26% -27%TxLDL 119 154 98 90 113 112

secondary high risk primary

% CHD

events on

statin

J Am Coll Card 2005;46:1225-8

LDLc lowering and residual risk – more is neededThe majority of CVD events still occur: CVD events occurring in the on-treatment groups in major statin trials

Despite the need beyond LDLc lowering, outcomes data supporting combination therapy still limited

ADA Standards of Care; Diabetes Care, January 2011

The lipid arm of the ACCORD trial was relatively disappointing for combination therapy (as was FIELD in 2005)– WHY?

April 29, 2010 N Engl J Med

Conclusion: “The combination of fenofibrate and simvatatin did not reduce the rate of fatal cardiovascular events, non-fatal MI or non-fatal stroke, as compared with simvatatin alone.”

ACCORD LIPID: Lipid parameters

ADA guidelines, 2007

ADA guidelines, 2008-2011

ADA guidelines changed text related to lipoprotein control in 2008:

Statins are safe but nothing is without risk: Review of 35 statin therapy trials

Kashani A et al. Circulation. 2006;114:2788-97.

FDA-approved statin* monotherapy vs placebo (N = 74,102)

*Atorvastatin, fluvastatin, lovastatin, pravastatin, rosuvastatin, simvastatinCK = creatine kinaseAE = adverse events

OutcomeStatin(%)

Placebo (%) RD P value

Myalgias 15.4 18.7 2.7 0.37

CK elevations 0.9 0.4 0.2 0.64

Rhabdomyolysis 0.2 0.1 0.4 0.13

LFT elevation 1.4 1.1 4.2 <0.01

AE discontinuation 5.6 6.1 -0.5 0.80

Statin better Placebo better

-30 -15 0 15 30Risk difference per 1000 patients (RD)

(95% CI)

20,536 patients from HPS (40mg simvastatin vs. placebo): 8 cases of rhabdomyolysis, 3 on placebo

Lancet 2002;360:7-23

Subjective myopathy incidence is high but myositis and rhabdomyolysis is low

Rhabdomyolysis per 10,000 person-years of therapy with lipid lowering agents*

Drug Monotherapy incidence +agent Combo. Incidence rates rates‡

Atorvastatin 0.54 fenofibrate 22.45

Cerivatatin 5.34 gemfibrozil 1035

Simvastatin 0.49 gemfibrozil 18.73

Gemfibrozil 3.70

*Data from 252,460 patients enrolled in 11 different health care plans in U.S. with at least 180 days enrolled in respective health care plan

‡ Data based on 7300 patient-years of combination therapy

Data from: [CDER] Graham D. JAMA 2004;292:2585-2590.

Use of run-in phases in recent, large trials of statin drugs results in likely under-estimation of myopathy incidence in practice

Control

Statin 40-80 mg

~ 5 years

Blinded Statin

4-8 weeks

Waters DD et al. Am J Cardiol. 2004;93:154-8.HPS investigators, Lancet 2002;360:7-22

1-8 weeks

Wash-out

Run-in

HPS: 63,603 patients 32,145 20,536TNT: 18,469 patients 15,464 10,003

Run-in: HPS, of 11,609 excluded after drug exposure, 1% for self-identifed ADRs

TNT, of 5,461 excluded after drug exposure, 3.7% for self- identified ADRs

Lipid summary:

Lipid management for cardiometabolic risk reduction:

• Use statin-based LDLc-lowering therapy to:• Achieve LDLc < 100 mg/dL in patients with diabetes 40 years of age and older with another risk factor• Achieve LDLc < 70 mg/dL in patients with diabetes and CVD

VI. Prevention and management of complications A. Cardiovascular disease 1. HTN control 2. Dyslipidemia management 3. Anti-platelet 4. Smoking cessastion

AJKD 2004;43(suppl 1):S120

Historic goal SBP of < 130 mmHg in diabetes is an extrapolation of data regarding benefits in nephropathy

July, 2009

Haven’t previous trials found a benefit from tighter BP control in diabetes?

…ended up comparing mean of 154/87 to 144/82

Cochrane review 2009

Four trials looked at major CVD outcomes based on randomized BP control;Two trials (ABCD) were exclusively in patients with diabetes

April 29th, 2010 N Engl J Med

ACCORD BP: Using an average of 3 drugs, the authors achieved a SBP of 119 mmHg vs. 133 mmHg

ACCORD BP: Results

Conclusions: “In patients with type 2 diabetes at high risk for cardiovascular events, targeting a systolic blood pressure of less than 120 mmHg, as compared with less than 140 mmHg, did not reduce the rate of fatal and nonfatal major CVD events.”

Implications on practice

Summary of the evidence: Lower BP goals

Lower BP goals: Do not change overall CV outcomes (all 3 trials). Do reduce rates of stroke (ABCD (H) and ACCORD, but how clinically sig?). Do help to reduce the progression of nephropathy in terms of urinary albumin excretion and

progression of microalbuminuria to overt albuminuria (ABCD (H) and (N)).

Trial Goal (mmHg) Achieved (mmHg)

ABCD (H)

DBP 75 vs 80-89 132/78 vs 138/86

ABCD (N)

DBP 10 < baseline vs 80-89

128/75 vs 137/81

ACCORD SBP <120 vs <140 119/64 vs 133/70

VI. Prevention and management of complications A. Cardiovascular disease 1. HTN control 2. Dyslipidemia management 3. Anti-platelet 4. Smoking cessastion

We have known for decades that platelets are more “responsive” in patients with diabetes. Reasons are still not fully understood nor the impact on use of anti-platelet agents

NEJM 2007;357:2482-94

Pignone M, Williams CD. Nature Review Endo, November 2010

The promiscuous platelet: many pathways exist for platelet activation. So do overactive platelets respond better or less well to standard anti-platelet therapy?

ASA: The benefit of anti-platelet therapy is greater in higher risk patients and quite low in low risk patients

Carlo Patrono, Barry Coller, Garret A. FitzGerald, Jack Hirsh, and Gerald Roth CHEST 2004;126: 234S-264S.

2 Events prevented per 1000 treated in healthy

population

Case for CVD prevention with ASA in moderate risk patients not so clear

FDA committee votes not to approve aspirin for the primary prevention of MI

Tue, 09 Dec 2003 21:00:00 Michael O'Riordan

Gaithersburg, MD - The evidence supporting the use of aspirin for the primary prevention of MI failed to hold up to the scrutiny of the FDA's Cardiovascular and Renal Drugs Advisory

Committee at its most recent December 8, 2003 meeting.

The committee voted overwhelmingly 11 votes against and three votes for approval of the petition sought by Bayer Corp to approve aspirin for the reduction of the risk of a first MI in moderate-

risk patients, those with a 10-year coronary heart disease risk >10%.Despite the existing data, which consisted of five major clinical trials, the committee felt the

evidence supporting the extended label for aspirin was inconsistent at best or lacking at worst.

Risk vs. benefit in primary vs. secondary prevention with ASAWhile the benefit of aspirin increases as risk increases, bleeding stays constant

So the benefits of antiplatelet therapy in low-risk patients is offset by major bleeding episodes:

NEJM 2005;353:2373-83

Diabetes Care, June 2010

Meta-analysis in general population for primary prevention:

2009 Update from Oxford group on clinical trials evidence for aspirin

Conclusion: “For primary prevention, aspirin is of uncertain net value as reduction in occlusive events is weighed against increase in major bleeds…….”“There is the possibility that some category of individual will eventually be identified in which primary prevention with aspirin is of definite net benefit. One particularly important such category might be adults with diabetes…..”

ASA placebo

Nine trial meta-analysis in ADA/AHA/ACCF statement:

CHD: RR 0.91 (0.79-1.05)Stroke: RR 0.85 (0.66-1.11)

What about bleeding in patients with diabetes?

Generic estimate ~ 1/1000 per year for non-stroke bleeding and ~ 1/10,000 for hemorrhagic stroke

In patient-level ATT meta-analysis, patients with diabetes examined separately: 25 GI bleeds with ASA (0.23%) and 22 bleeds with placebo (0.21%)Hemorrhagic stroke: 6 events on ASA, 9 on placebo

The Bottom Line

At a 10% 10-year risk of MI and Stroke, aspirin would prevent 1 MI and 1 stroke and maybe cause 1 major GI bleed. At a 20% 10-year risk, 2 MIs and 2 strokes would be prevented with no change in bleed risk

ADA Standards of Care, Diabetes Care; January, 2011

Conclusions:

For cardiometabolic risk reduction:

• A lot of new clinical trial data has solidified current clinical goals:• A1C ~ 7% • SBP ~ 130 mmHg• LDLc < 100mg with statin (ACCORD lipid achieved 80 mg/dL)• Use of ASA in higher risk patients with diabetes

• Same data suggests that more aggressive targets not warranted for CVD reduction:• A1C < 7%• SBP < 130 (ACCORD achieved 119 mmHg SBP)• Addition of fibrate or niacin to statin to target TG if LDLc already at goal• ASA should not be used if CVD risk is low