Treating arteries instead of treating risk factors

J. David Spence Stroke Prevention & Atherosclerosis Research Centre

Robarts Research InstituteLondon, Canada

dspence@robarts.ca www.robarts.ca/sparc

Disclosures• Grants for research from HSF, NIH, CIHR• Grants for research from Pfizer, Merck, Pan

American Labs• Lecture fees from Pfizer, AstraZeneca, Merck,

Novartis, Boehringer-Ingelheim• Consulting fees from Novartis, Boehringer-

Ingelheim• Interest in www.vascularis.com

Post-prandial oxidative stress and inflammation*

* ROS, inflammatory mediators, oxidized LDL: not fasting Chol/Trig/HDL

Large artery strokes• Not just stenosis: also high plaque burden• Plaque measurement very useful

79 yo woman

Composite drawing of all plaques in extracranial carotidsBogiatzi C…Spence JD SPARKLE classification Neuroepidemiology 2014;42:243–251.

72 yo man

Ischemic stroke subtypes are changing• Better BP control• More statins

Bogiatzi C ….Spence JD. Stroke. 2014 Sep 11

Ischemic stroke subtypes are changing

Bogiatzi C ….Spence JD. Stroke. 2014;45:3208-13.

Cardioembolic strokes more common, large artery strokes less common

Before 2005 After 2009

Measurement of subclinical atherosclerosis

• There are 2 distinct kinds of IMT• IMT isn’t atherosclerosis• Plaque predicts events better than IMT• Plaque measurement can be used for treatment• Plaque measurement is more sensitive to effects

of therapy• Plaque measurement is superior to IMTSpence JD. Atherosclerosis. 2012;220:34-5.

Ultrasound measurement of “Atherosclerosis”

It is important to recognize biological differences among

•Intima-media thickness - with and without plaque thickness•Plaque•Stenosis

Carotid Intima-Media Thickness (IMT)Mannheim consensus conference– Site : common carotid artery , far wall ,– Quality Index > 0.50

Cerebrovascular Diseases 2007;23:75-80

Phenotypes of atherosclerosis

Traditional coronary risk factors as predictors of ultrasound phenotype:

In multivariable regression: • IMT R2 is 0.15 for internal, 0.17 for common carotid1

• Plaque area R2 is 0.522 (similar to R2 for coronary events)• Stenosis (Doppler velocity) R2 is 0.132

1. O’Leary DH, et al Stroke 1996; 27: 224-231.2. Spence JD, Hegele RA Stroke 2004; 35: 649 - 653.

Scannning and plaque area measured by Maria DiCicco RVT; IMT by Andrew House M.D., plaque volume by Khalid Al-Shali M.D.

IMT=0.95 mm PA=0.38 cm2 PV=297 mm3

IMT=1.01 mm PA=0.19 cm2 PV=24 mm3

Measurement of IMT, plaque area and volume

Scannning and plaque area measured by Maria DiCicco RVT; IMT by Andrew House M.D., plaque volume by Khalid Al-Shali M.D.

IMT=0.95 mm PA=0.38 cm2 PV=297 mm3

IMT=1.01 mm PA=0.19 cm2 PV=24 mm3

Al-Shali et al. Atherosclerosis 2005; 178: 319–325

Scannning and plaque area measured by Maria DiCicco RVT; IMT by Andrew House M.D., plaque volume by Khalid Al-Shali M.D.

IMT=0.95 mm PA=0.38 cm2 PV=297 mm3

IMT=1.01 mm PA=0.19 cm2 PV=24 mm3

Phenotypes of atherosclerosisThus Intima-media thickness (IMT), plaque and stenosis must be

regarded as distinct phenotypes, with distinct biologies and determinants. Therapies would also be expected to differentially affect these distinct manifestations of atherosclerosis.

Biologies of IMT, plaque and stenosis are distinctIMT: mainly hypertensive medial hypertrophyPlaque: reflects endothelial dysfunction, oxidative stress, lipidsStenosis: consequence of plaque rupture and thrombosis – reflects plaque instability, inflammation, MMP, thrombosis, impaired

fibrinolysisSpence JD, Hegele RA Noninvasive Phenotypes of Atherosclerosis: Similar Windows but Different Views Stroke

2004; 35: 649 - 653.Spence JD, Hegele RA. Noninvasive phenotypes of atherosclerosis. Arterioscler Thromb Vasc Biol. 2004

Nov;24(11):e188

Plaque is more closely related to coronary artery disease than IMT

• Ebrahim S, et al. Stroke 1999; 30:841-850.

• Chan SY et al. J Am Coll Cardiol. 2003;42:1037-43.

• Brook R et. al. ATVB 2006;26:656-62.

• Johnsen, SH et al. Stroke 2007;38;2873-2880

• Inaba Y, et al. Atherosclerosis. 2011;220:128-33.

IMT isn’t atherosclerosisCorrelation Between Carotid Intimal/Medial Thickness (IMT) and

Atherosclerosis: A Point of View from Pathology Finn AV, Kolodgie FD, Virmani R. ATVB 2009 online

Measurement of 2-D Plaque area*

* Invented in our lab in 1990 by Maria DiCicco, R.V.T.

Carotid Plaque Area as predictor of events

1,686 patients in our Atherosclerosis Prevention Clinic followed up to 5years

During mean followup of 2.5 + 1.3 years:94 MI, 45 strokes, 44 deaths (27 vascular).

Spence JD, Eliasziw M, DiCicco M et al. Carotid Plaque Area: A Tool for Targeting and Evaluating Vascular Preventive Therapy. Stroke. 2002;33:2916-2922.

Baseline Carotid plaque area as a predictor of eventsStroke, MI, Death (after adjustment for risk factors*)

*Age, sex, SBP, tChol, pack-yrs, tHcy, diabetes, Rx lipids and BPStroke 2002; 33:2916-2922.

Prediction of outcomes

Plaque measurement is a stronger predictor of outcomes than EBCT, presence of plaque, and somewhat more predictive than IMT, particularly for myocardial infarction

TPA increases AUC in ROC

Romanens M, Spence JD et al. Cardiovasc. Med. 2011;14:53–57

Tromsø Study

6226 men and women aged 25 to 84 6 year followup: • MI in 6.6% of men and 3.0% of women. • TPA: RR (95% CI) 1.56 (1.04 to 2.36) in men 3.95 (2.16 to 7.19) in women• IMT RR (95% CI) 1.73 (0.98 to 3.06) in men

2.86 (1.07 to 7.65) in women• When bulb IMT was excluded from analyses, IMT

did not predict MI in either sex.Johnsen, SH et al. Stroke 2007;38;2873-2880

5-year MI risk by Total Plaque Area Tertile

Johnsen, SH et al. Stroke 2007;38;2873-2880

Men Women

IMT in the CCA was not predictive

10-year stroke risk more strongly predicted by plaque area in Tromsø Study

Mathiesen ES et al. Stroke 2011 online Feb 10

Hazard ratio 1.73 for men(p=0.004), 1.63 for women (p=0.03)No differences for quartiles of IMT

Total plaque area

Distribution of carotid plaque area by age groups and sex

Spence JD. Nature Clinical Practice Neurology 2006;2: 611-619.

Plaque progression despite therapy doubles the risk*

*Adjusted for Age, sex, SBP, tChol, pack-yrs, tHcy, diabetes, Rx lipids and BPStroke 2002; 33:2916-2922.

Medical treatment was failing in half the cases, and they were at double the risk: we needed to do better!

Paradigm change:Treating arteries, not risk factors

Instead of treating risk factors to target, since 2003 we treat patients more intensively if their plaque is progressing , regardless of their level

of LDL or other risk factors

i.e. – since 2003 our target is now plaque regression

Treating arteries without measuring plaque is like

treating hypertension without measuring blood pressure

Benefit of carotid endarterectomy

Symptomatic severe stenosis: 2-yr reduction of stroke death from 26% to 9%Asymptomatic: 5-yr risk reduction 10% to 5% NNT to prevent 1 stroke in 2 years1:

NNT

Symptomatic severe >70% age<75 6

Symptomatic severe >70% age>75 3

Symptomatic moderate 50-69% 15

Asymptomatic 67-83* Predicated on 3% surgical risk, and historical medical therapy1.Barnett HJM. CMAJ 2004;171: 473-4

TCD microembolus detection

319 ACS patients between 2000 and 2004

10% had microemboli

1-year Stroke RiskNo Emboli Emboli 1% 15.6%

95% CI (1.01 -1.36) (4.1-79)

p<0.0001

Spence JD et al. Stroke 2005; 36:2373-2378.

Stroke risk over 2 years by baseline microembolic status

0 200 400 600 800

Days to event

0.0

0.2

0.4

0.6

0.8

1.0

Su

rviv

al fr

ee

of

str

ok

eEmboli at baseline

No

Yes

0-censored

1-censored

Spence JD et al. Stroke 2005;36:2373-2378

11%

2.2%

P<0.001

Spence JD et al. Arch Neurol. 2010;67:180-6

Decline of microemboli with more intensive medical therapy

< 5% of ACS patients can now benefit from carotid endarterectomy or stenting

Annual rate of plaque progression in ACS patients before and since 2003

Spence JD et al. Arch Neurol. 2010;67:180-6

n= 346P<0.0001

Kaplan-Meier Survival free of stroke, death, MI

logrank test p<0.0001 logrank test p<0.0001

Spence JD et al. Arch Neurol. 2010;67:180-6

Plaque area by age group and clinic pop. age by year

n= 4,328Spence JD, Hackam DG. Stroke 2010 Jun;41(6):1193-9

n= 4,328

Average rate of plaque progression by year among all patients in clinic

Effectsof

more intensivetherapy

on plasma

lipidsin clinic

populationn=4,328

Rates of progression and LDL by year

1997- 1998

1998- 1999

1999-2000

2000- 2001

2001-2002

2002-2003

2003-2004

2004-2005

2005-2006

2006-2007

Age at first year (SD)

50.45 (11.34)

54.96 (13.24)

57.43 (13.09)

60.54 (12.37)

61.33 (12.37)

62.88 (13.05)

63.50 (12.74)

63.70 (12.51)

64.02 (13.13)

64.72 (13.50)

Progression 28.2% 33% 48.1% 61.7% 55.4% 46.1% 41.9% 40.6% 26.8% 28.4% Stable 35.9% 28% 23.3% 18.1% 18.5% 20.4% 20.1% 25.6% 23.1% 31% Regression 35.9% 38.7% 28.6% 19.6% 26.1% 33.4% 38% 33.9% 50.1% 40.5%

LDL 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Progression 3.61 (2.33)

2.45 (0.95)

2.29 (0.77)

2.43 (0.81)

2.35 (0.82)

2.34 (0.73)

2.32 (0.81)

2.22 (0.55)

2.14 (0.83)

1.84 (0.74)

Stable 2.93 (0.74)

2.61 (0.57)

2.21 (0.79)

2.58 (0.74)

2.41 (0.89)

2.51 (0.98)

2.34 (0.74)

2.35 (0.89)

2.12 (0.81)

2.25 (0.93)

Regression 2.63 (1.02)

2.20 (0.32)

2.18 (0.93)

2.37 (0.77)

2.38 (0.76)

2.42 (0.95)

2.25 (0.64)

2.01* (0.61

1.94 (0.83)

1.87 (0.78)**

Spence JD, Hackam DG. Stroke 2010 Jun;41(6):1193-9

Decline in events in ACS with more intensive medical therapy

No embolin=431

Microembolin=37

pBefore 2003n=199

Since 2003n=269

p*

Stroke in year 1 1.4% 10.3% 0.016 3.3% 1% 0.155

Stroke in year 2 1.8% 18.5% 0.001 5.5% 0% 0.006

MI in year 1 2.2% 6.9% 0.165 4.9% 0.5% 0.007

MI in year 2 1.4% 3.2% 0.394 2.7% 0.5% 0.104

Death in year 1 2.8% 10.3% 0.069 4.4%% 2.4% 0.386

Death in year 2 2.1% 3.7% 0.477 3.8% 0% 0.044

CEA year 1 1.4% 12.9% 0.003 2.7% 1.9% 0.739

CEA year 2 0.3% 3.7% 0.146 1.1% 0% 0.499

Stroke, death or CEA 1st 2 years

6.5% 32.4% <0.0001 14.1% 4.5% <0.0001

Stroke, death, MI or CEA 1st 2 years

8.6% 32.4% <0.0001 17.6% 5.2% <0.0001

Spence JD et al. Arch Neurol. 2010;67:180-6

Carotid plaque measurementSummaryPlaque measurement is useful for:

Managing patientsStratifying riskManaging resourcesEncouraging patients to follow regimenMonitoring success of therapy

Genetic researchQuantitative traits for linkage studies

Studying effects of new therapiesMuch smaller sample size x durationProof of concept studies in human subjectsDose-finding studies

Plaque measurement to study effects of new therapies• New therapies being developed for atherosclerosis - effective in animal models - no effect on blood pressure or lipids• It will be necessary to measure plaque - for dose finding studies - to demonstrate efficacy before committing to

very expensive events –based studies

Eg: inhibitors ACAT CETP, Leukotriene B4, etc.

Sample size x duration required

To show a 30% reduction in rate of progressionPower 80%, p<0.05

IMT: 468 patients/group x 2 years1

(less with automated edge detection)

Plaque area: 75-100 patients /group x 2 years2

3-D plaque volume: ?1. Bots M et al, Stroke 2003;34:2985-2994 2. Hackam DG et al Am J Hypertens 2000;13:105-10.

Disk segmentation

for measurement

of plaque volume

Disk segmentation method

First measurement of carotid plaque volume 1994

Rendered plaque volume

Plaque volume fixed at bifurcation

Stroke 2005; 35: 1904-1909.

Plaque volume fixed at bifurcation

3-D ultrasound carotid plaque volume: a tool for quickly measuring effects of treatment on atherosclerosis

38 patients with carotid stenosis >60% age 68 ± 6.6 years, 15 female, randomly assigned to atorvastatin 80mg daily

(n=17) vs placebo (n=21)

Stroke 2005; 35:1904-1909.

Rate of plaque volume progression on placebo vs Atorvastatin 80mg

In 3 months:Placebo Atorvastatin 16.8 + 74.1 mm3 -90.24 + 85.12

mm3 (p<0.0001)

Stroke 2005; 35:1904-1909.

Placebo Atorvastatin

Treatment group

-100.00

-50.00

0.00

Ch

an

ge in

pla

qu

e v

olu

me (

mm

3)

+-

SE

P<0.0001

Stroke 2005; 35:1904-1909.

3-month progression of carotid plaque volume with placebo vs. atorvastatin 80mg

To show treatment effect in 3 months placebo progression 16.81 + 80 mm3

Sample size for change in progression of plaque volume

Power Effect size (% of atorva)

Regression in mm3

Sample size Per group

.9 10% -9 203

.9 25% -23 86

.9 50% -45 36

.9 75% -68 20

Sample size for change in progression of plaque volume

To show treatment effect in 6 months assuming linear progression on placebo and regression on

active treatment; placebo progression 33.62 + 80 mm3

Power Effect size (% of atorva)

Regression in mm3

Sample size Per group

.9 10% -18 51

.9 25% -45 22

.9 50% -90 9

.9 75% -135 6

Stroke 2005; 35:1904-1909.

Vessel Wall Volume

Egger M, Spence JD, Fenster A, Parraga G. Validation of 3D Ultrasound Vessel Wall Volume: An Imaging Phenotype of Carotid Atherosclerosis. Ultrasound Med Biol. 2007 Jun;33(6):905-14.

Atorvastatin 80 vs placebo on VWV

VWV Placebo Atorvastatin p

+70 ± 140 mm3 -30 ± 110 mm3 <0.05

(14.9 ± 10.3%) (-1.4 ± 7.7%)

Krasinski A, Chiu B, Spence JD, Fenster A, Parraga G. Ultrasound Med Biol. 2009 Jul 31.

DIRECT 3D Ultrasound Study

Shai I, Spence JD, Parraga A, Mallette C, Fenster JD et al. Circ 2010;121:1200-1208.

Atherosclerosis regression on all 3 diets, proportional to BP reduction and weight loss

Annual change cannot be measured within patients by IMT

Resolution of carotid ultrasound is ~ 0.3mm

Mean rate of change of IMT is only 0.0147 for mean and 0.0176 for maximum IMT1

Large groups required to show changes for groups, not individuals

Mean change in TPA is 11mm2; can easily be measured.

1. Bots ML, et a. Stroke 2003 Dec;34(12):2985-94.2. Spence JD. Can J Cardiol 2008; 24 (Suppl C): 61C-64C.

11+ 34 mm2

Plaque measurement is superior to IMT

• Greater dynamic range (~ 100-fold)• More predictive of stroke and MI• More sensitive to effects of therapy

Spence JD. Plaque measurement is superior to IMT. Atherosclerosis 2011.

Treating arteries without measuring plaque is like

treating hypertension without measuring blood pressure

3D Volume Analysis

Measurement of plaque volume at baseline, and change over time

Enable Imaging Technologies Beijing

Acknowledgements3-D Ultrasound technology GeneticsDrs. Aaron Fenster, Grace Parraga Dr. Rob Hegele Measurements Lab Manager2-D : Maria DiCicco RVT Tisha Mabb3-D: Craig Ainsworth, Funding Anthony Landry, Chris Blake, NINDSMicaela Egger, Christiane Mallet, Silvia Riccio HSF Ontario Bernard Chiu, Shayna McKay, Adam Krasinsky CHRI Ulcers Dr. Vadim Beletsky, Jeremy Mason Plaque composition Jeremy Mason, Dr. Joseph AwadTCD Study Dr. Arturo Tamayo MRIDr. Claudio Munoz Dr. Brian RuttScanning PET/CTMaria DiCicco RVT Dr. Jean-Luc Urbain Janine Desroches RVT Dr. Ting Lee

Acknowledgements

Maria DiCiccoR.V.T.Plaque area

Aaron FensterPh.D.Plaque volumePlaque roughness, texture3D U/S

Grace ParragaPh.D.Vessel wall volume

http://www.robarts.ca/sparc dspence@robarts.ca