Regression An Overview of Recent Findings & Issues

13th Angioplasty Summit 2008 Cheol Whan Lee, MD University of Ulsan, Asan Medical Center, Seoul, Korea

CardioVascular Research Foundation Asan Medical Center Atherosclerosis

Progression

Compensatoyr expansion Expansion overcome maintains constant lumne lumen narrows Reversibility of Arterial Lesions

Animal models: documented in a wide range of species

Humans:Normal Minimal Moderate Severe Therevessel are growingCAD evidencesCAD of plaqueCAD regression in mild atherosclerotic disease Regression

* Atherosclerosis is usually viewed as a chronic progressive disease characterized by continuous accumulation of atheroma within the arterial wall. Presentation

• Evidences of plaque regression • Predictors of plaque regression • Clinical significance of regression • Future perspective Evidences of Plaque Regression Carotid Artery Effect of on Early Carotid Atherosclerosis & CV Events: ACAPS

0.03

Placebo Lovastatin (20-40 mg/d) 0.02 versus placebo

0.01 919 asymptomatic patients with early 0.00 carotid atherosclerosis

Relative Mean Relative Mean Primary outcome: -0.01 Lovastatin 3-year change of

Maximum Thickness (mm) Maximum Thickness mean maximum IMT -0.02 Baseline 6,12 Mo 18, 24 Mo 30, 36 Mo In men & women with moderately elevated LDL , lovastatin reverses progression of IMT in the carotid arteries & appears to reduce the risk of major CV events & mortality. Circulation 1994;90:1679 ASAP 0.09 80mg, n=160 0.07 40 mg, n=165 LDL-C>174 mg/dl Aggressive LDL- 0.05 Ica (S) Bul (S) Overall (S) cholesterol reduction 0.03 was accompanied by 0.01 regression of carotid -0.01 Cca (S) intima media thickness Bul (A) -0.03 Ica (A) in patients with familial Overall (A) -0.05 Cca (A) hypercholesterolaemia,

Change in thickness (mm) Change in thickness whereas conventional -0.07 LDL lowering was not. -0.09 0 1 2 Years Lancet 2001; 357: 577-81 The puzzling result different direction ENHANCE

0.80 a double-blind, randomized, 24-month trial comparing the effects of daily therapy with 80 mg of simvastatin either with placebo or with 10 mg of in 720 patients with familial .

0.75 LDL-C 16.5%↓ CRP 26%↓ P=0.88 HDL 3%↑ 0.70 Mean IMT (mm) 0.65 Simva 80mg (n=363) Eze10-Simva 80mg(n=357) 0.60 6 12 18 24 Months

In patients with FHC, combined therapy with ezetimibe & simvastatin did not result in a significant difference in changes in IMT, as compared with simvastatin alone.

NEJM 2008 on line METEOR

+0.03 Placebo +0.0131 mm/yr (n=252) +0.02

P<0.0001 +0.01 (CRESTOR vs. placebo)

1 2 Time (years) 0.00

P=NS (CRESTOR vs. zero slope 40 mg -0.01 -0.0014 mm/yr Change in IMT of 12 carotid sites (mm) Change in IMT of 12 carotid sites (n=624)

In middle-aged adults with an Framingham risk score of less than 10% & evidence of subclinical atherosclerosis, rosuvastatin resulted in statistically significant reductions in the rate of progression of maximum CIMT over 2 years vs placebo.

Crouse JR, et al. JAMA 2007;297(12):1344 SummarySummary ofof MajorMajor LipidLipid--ModifyingModifying TrialsTrials Trials Patients No/ Treatment Target Difference Duration (y) (mm/year) ACAPS 919/3 L20-40/Pl IMT -0.015 ARBITER1 161/1 A80/P40 IMT -0.059 ASAP 325/2 A80/S40 IMT -0.033 CAIUS 305/3 P40/Pl IMT -0.0132 ENHANCE S80+E1/S80 IMT 0.0026 KAPS 424/3 P40/Pl IMT -0.014 LIPID 522/4 P40/Pl IMT -0.0155 METEOR 984/2 R40/Pl IMT -0.0145 ORION 43/2 R80/R5 Volume No change PLAC-II 151/3 P20-40/Pl IMT -0.0082 RADIANCE1 850/2 T60+A/A IMT -0.0006 RADIANCE2 752/1.8 T60+A/A IMT 0.0050 A: atorvastatin, E: ezetrol, L: lovastatin, P: , Pl: placebo, R: rosuvastatin, S: simvastatin, T: torcetrapib Evidences of Plaque Regression Aorta Effects of Simvastatin on Atherosclerotic Lesions: MRI Study

N=18 LDL-C≥130 mg/dL, TG≤445 mg/dL

Effective and maintained lipid-lowering therapy by simvastatin is associated with a significant regression of atherosclerotic lesions (reduction in lipid content -> plaque stabilization).

Circulation. 2001;104:249 Different Susceptibilities of Thoracic and Abdominal Aortic Plaques to Lipid Lowering Thoracic Aorta Abdominal Aorta The effects of 20-mg versus 5-mg atorvastatin on thoracic and abdominal aortic plaques in 40 hypercholesterolemic VWA VWA patients (MRI) for 1year.

Regression of thoracic aortic plaques, whereas only retardation of plaque progression in abdominal aorta.

LDL-C LDL-C

JACC2005:45:733 Evidences of Plaque Regression Coronary Artery REVERSAL Double-blind period

Atorvastatin 80 mg/day Placebo Screening Randomization Run-in Visit* Phase 654 patients Pravastatin 40 mg/day

*Includes baseline intravascular ultrasound (IVUS) 18-month follow-up with IVUS

JAMA 2004;291:1071 REVERSAL ) ) ) 20 3 3 3

15 Pravastatin 40 mg (n=253) 10

5

0 Atorvastatin 80 mg -5 (n=253)

-10 Change in Atheroma Volume (mm Change in Atheroma Change in Atheroma Volume (mm Change in Atheroma Change in Atheroma Volume (mm Change in Atheroma -15 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 Change In LDL Cholesterol (%) Intensive lipid-lowering treatment with atorvastatin reduced progression of coronary atherosclerosis compared with pravastatin. The progression rate at any level of LDL-C reduction was lower with atorvastatin compared with pravastatin. ASTEROID

To evaluate whether long-term treatment with rosuvastatin 40 mg in CAD pts resulted in coronary plaque regression (single arm study)

Patients CAD, undergoing coronary angiography Target coronary artery: ≤50% Rosuvastatin 40 mg reduction in lumen diameter of (n=349 evaluated serial IVUS examinations) ≥40 mm segment No cholesterol entry criteria ≥18 years Visit: 1 2 3 4 5 6 7 8 9 10 0 13 26 39 52 65 78 91 104 Week: –6

Eligibility IVUS Lipids TolerabilityLipids Tolerability Tolerability IVUS assessment Lipids Tolerability Tolerability Lipids Tolerability Tolerability Primary Endpoint: ∆% Atheroma Volume Median Percent Atheroma Volume 0

Atheroma Area 10.16 mm2 -0.1 -0.2

Lumen Area 6.19 mm2 -0.3 -0.4 -0.5 -0.79% -0.6 Atheroma Area 5.81 mm2 (64% of pts) -0.7

Lumen Area (% baseline from change 5.96 mm2 -0.8 -0.9

Very high-intensity therapy using rosuvastatin 40mg/d resulted in significant regression of atherosclerosis, Much heat, little light Effect of Recombinant ApoA-I Milano (HDL Mimetics) on Coronary Atherosclerosis in Pts With ACS

-14.1 mm3 or a 4.2% decrease from baseline A recombinant ApoA-I Milano/phospholipid complex (ETC-216) administered intravenously for 5 doses at weekly intervals produced significant regression of coronary atherosclerosis as measured by IVUS. JAMA2003:290:2292 TheThe evidenceevidence

SummarySummary

There are convincing evidences to support that high-dose statin or HDL-raising therapies may induce regression of mild to moderate atherosclerotic lesions in humans. Presentation

• Evidences of plaque regression • Predictors of plaque regression • Clinical significance of regression • Future perspective More Aggressive Targets for LDL & Blood Pressure Effect of Lower Targets for BP & LDL Cholesterol on Atherosclerosis in DM The SANDS Randomized Trial

The probability of a 0.04 499 american indian age>40y & diabetes decrease in IMT was 0.03 significantly related to 0.02 decrease in LDL-C but not significantly related 0.01 p<0.001 to a decrease in SBP.

0 Conversely, probability -0.01 of decreases in LVMI were significantly IMT/ 36 months Aggressive Standard -0.02 related to decreases in ∆ (n=252) (n=247) -0.03 SBP but not to LDL-C decreases. -0.04

Reducing LDL-C (<70mg/dl vs.100 ) & SBP (<115mmHg vs. 130) to lower targets resulted in regression of carotid IMT & greater decrease in LV mass in type 2 DM.

JAMA2008;299:1678 REVERSAL: Continuous Relationship between LDL-C, CRP, and Percent Atheroma Volume

Change In Percent Change In Percent Atheroma Volume (%) Atheroma Volume (%) 3.5 3.5

3 2.5 2.5 In REVERSAL, 1.5 2 0.5 Greatest decrease in disease1.5 progression observed -0.5 among patients with greatest1 reductions in CRP for any given change in0.5 LDL-C -1.5 0 -2.5 -100 -75 -50 -25 0 25 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 Change In LDL-C (mg/dL) Change In CRP (mg/L)

Adapted from Nissen et al. N Engl J Med. 2005;352:29 Usefulness of Follow-Up LDL–C Level as an Independent Predictor of Changes of Coronary Atherosclerotic Plaque Size After Statin Therapy

N=103, 1 year follow-up

When patients achieved a follow-up LDL cholesterol level <100 mg/dl, regression or no progression of coronary plaque was expected.

Hong MK et al, AJC 2006;98:866 Relationship between LDL-C levels &change in % atheroma volume for several IVUS trials

1.8 REVERSAL R2 = 0.97 pravastatin P<0.001 CAMELOT placebo 1.2 MedianIn ASTEROID, changeThere in was no plaque regression in patients with LDL-C>100mg/dl, Percent 0.6 REVERSAL Atheroma atorvastatin Volume and little change in patientsA-Plus with LDL-CProgression of 70-100mg/dl. (%) placebo Plaque regression0 was only seen in patients with LDL-C≤ 70mg/dl. Regression -0.6 ASTEROID rosuvastatin -1.2 50 60 70 80 90 100 110 120 Mean LDL-C (mg/dL)

JAMA 2006; 295(13):1556 TheThe regressibleregressible plaqueplaque

SummarySummary

• Plaque regression is associated with a substantial reduction of LDL-C, and lipids and inflammatory substances may be responsible for this change.

• It remains uncertain whether plaque consisting of connective tissue and calcium may be regressible. Presentation

• Evidences of plaque regression • Predictors of plaque regression • Clinical significance of regression • Future perspective Association between Carotid IMT and Clinical Endpoints

Are surrogate imaging markers of the arterial r= -0.7, p=0.01 wall are representative for clinical outcomes ?

Imaging trials and clinical endpoints trials that evaluated similar drugs indicates good correlations between CIMT & clinical events.

Limitations: - different imaging protocols - different outcomes measures

Kastelein et al. Curr Opin Lipidiol 2007;18:613 The avandia debate Effect of Rosiglitazone on Carotid IMT Progression in CAD Patients Without DM Placebo Rosiglitazone 8 mg/day

Progression rate = 0.031 mm/48 weeks Progression rate = -0.012 mm/48 week (0.0016, 0.0604) (-0.0414, 0.0174)

0.04 0.04 0.03 0.03 0.02 0.02 0.01 0.01 0 0

IMT change (mm) -0.01 -0.01 0 24 48 0 24 48

Time (weeks) Time (weeks) Rosiglitazone reduces common carotid IMT progression in nondiabetic CAD patients, & insulin-sensitization may be one contributory mechanism. ATVB 2004;24:930 An effective surrogate?

Estrogen in the Prevention of Atherosclerosis A Randomized, Double-Blind, Placebo-Controlled Trial

0.01

0.005 P<0.05

0 IMT/year

∆ -0.005 Estradiol Placebo (n=97) (n=102)

-0.01

The average rate of progression of IMT was slower in healthy postmenopausal women taking unopposed ERT with 17ß-estradiol than in women taking placebo Ann Intern Med 2001;135:939 A very different interpretation ENHANCE vs. ASAP What’s the differences? A seemingly perfect analogue 0.04 LDL-C differences: 0.03 ∆ENHANCE: 17% P=NS Simva 0.02 40 mg ASAP: 9% 0.01 P<0.01 Not the end of the world! - a surrogate endpoint 0 - IMPROVE-IT The results Simvaof IMPROVE-IT 80 Simva 80 , which will not be available untilDifferences 2012, are expected

IMT (mm) IMT -0.01 Ezetrol 10 - thinner (0.7 vs. 0.9mm) Atorva - plaque lipid depletion: not∆ -0.02 only to help define the role of ezetimibe in the80 treatment mg of hypercholesterolemiastatin pretreatment (80%) but also to provide insight into the use of IMT as a surrogate indicator- trigger of proatherogenic coronary events. genes -0.03 -0.04 ENHANCE ASAP Association between Coronary Plaque Progression as Measured by IVUS & CV Events

The REVERSAL study used the same treatment regimen as the PROVE-IT. Although the REVERSAL & PROVE-IT studies were distinct studies, their results provided evidence that plaque progression measured by IVUS is predictive of an increased risk of CV events. The same direction Stunning reversals

Torcetrapib & Regression ILLUSTRATE

Placebo torcetrapib ) 3 0 N=1,188 % atheroma volume IVUS-derived indexes of coronary plaque0.19 progression vs 0.12 (p=0.72) -6.3 -9.4 AV of most diseased 10 mm -3.3 vs -4.1 (p=0.12) Pitfall of PAV (average atheroma area/average EEM area) -5 -5%↑ of atheroma (progression), 10%↑ of EEM HDL(positive 61% remodeling)↑ → 4.5%↓ of PAV (regression) LDL 20% ↓ It remains uncertain what’s p=0.02the best surrogate for 2/4mmHgclinical benefits.↑

-10 Total Atheroma Volume (mm

NEJM2007;356:1304 Clinical benefits vs. surrogate goals SummarySummary

• The CIMT or IVUS studies provide proof of concept, giving us the answer with only a few hundred patients

• Although plaque regression does not guarantee a reduction in the rate of clinical events, it may not be expected without a reduction in plaque progression.

• However, imaging surrogates will not be a substitute for clinical endpoint trials because regression does not necessary mean plaque stabilization. Presentation

• Evidences of plaque regression • Predictors of plaque regression • Clinical significance of regression • Future perspective EffectEffect ofof RosuvastatinRosuvastatin TherapyTherapy onon CoronaryCoronary ArteryArtery StenosisStenosis AssessedAssessed byby QQCACA inin ASTEROIDASTEROID

Clinically Relevant Changes

Stenosis reduced by ≥ 10% (regression*) 22 7.5%

Stenosis changed by < 10% 261 89 .4%

Stenosis increased by ≥ 10% (progression*) 93.1%

* Proportion of regressors greater than progressors, both p <0.03

MLD larger by ≥ 0.2mm (regression*) 34 12.1%

MLD Change < 0.2mm 261 89 .4%

MLD smaller by ≥ 0.2mm (progression*) 93.1%

* Proportion of regressors greater than progressors, both p <0.03

Circulation 2008;117: on line Paradoxical lumen loss

Minimal Regression Is Exchanged for Lumen Loss. Progression of coronary atherosclerosis can be associated with a paradoxical increase in lumen cross- sectional area, whereas regression is not associated with any change in lumen area (REVERSAL trial).

Atherosclerosis 2006;189(1):229 More than just regression! Future Perspective and Unanswered Questions

• Study duration: - Most studies lasted for only ≤ 2 years

• Target lesions: - Most studies targeted mild disease (DS 20~50%) - ThePlaque IVUS estimate Quantity is based vs. onQuality only 1 of the 3 major coronary arteries (partial sampling). - Minimal cosmetic improvement does not improve myocardial ischemia. - Plaque• Study regression endpoints: does not necessary mean plaque stabilization, requiring- Most imaging studies tools evaluated for accurate the changeevaluation of ofplaque plaque volume quality. - Plaque regression was minimal in most studies. Plaque regression, a good but not great surrogate

ConclusionsConclusions

Atherosclerosis progression can be •Regression is a surrogate marker (unproven stand-in)slowed andwaiting potentially for definitive reversed. outcome trials.

•AnyRegression drug taken is a by useful patients surrogate every day marker forin thean earlyrest of phase their oflife drug should development be backed with acceptedbefore clinical outcome endpoint data (Level trials. A evidence).