Jennifer G. Robinson, MD, MPH, Program Chair

1
An Examination of the ENHANCE Trial
Jennifer G. Robinson, MD, MPH, Program
Chair President, Midwest Lipid Association Associa
te Professor Departments of Epidemiology and
Medicine Director, Lipid Research
Clinic University of Iowa Iowa City, IA
2
Kastelein JJP, et al. N Engl J Med
20083581431-1443
3
ENHANCE Study Design
Pre-randomization Phase
RANDOMI ZAT I ON
FH LDL-c 210 mg/dL
Screening and Fibrate Washout
Placebo Lead-In/ Drug Washout
IMT assessment
-6
-10 to -7
0
21
24
3
18
6
9
12
15
Weeks
Months
4
ENHANCE Baseline Characteristics
Simvastatin Monotherapy Simvastatin plus Ezetimibe
All randomized patients n363 n357 P-value
Age (yr) 45.7?10.0 46.1?9.0 0.69
Male sex no. () 179(49) 191 (54) 0.26
Body-mass index 26.7?4.4 27.4?4.6 0.047
History of diabetes 5(1) 8 (2) 0.38
Hypertension 51 (14) 67 (19) 0.09
Current smoking 104 (29) 102 (29) 0.98
History of MI 26 (7) 14 (4) 0.06
Prior use of statins 297 (82) 286 (80) 0.56
Systolic mm Hg 124?15 125?15 0.31
Diastolic mm Hg 78?10 78?9 0.41
5
LDL-cholesterol
10
0
-10
-20
-30
Plt0.01
Percentage change from baseline
-40
-16.5 incremental reduction
-50
-60
-70
0
6
18
24
12
Months
ENHANCE
6
Primary outcome

• Change from baseline in mean CIMT
• Average of means of the far-wall IMT of R L
  carotid arteries, carotid bulbs, internal
  carotid arteries

ENHANCE
7
No significant changes in 1 or 2 endpoints
Variable Variable Simvastatin 80 mg Simvastatin 80 mg Simvastatin 80 mg Simvastatin 80 mg Ezetimibe 10 mg Simvastatin 80 mg Ezetimibe 10 mg Simvastatin 80 mg Ezetimibe 10 mg Simvastatin 80 mg Ezetimibe 10 mg P value(mean) P value(mean)
Mean Median Median Mean Median Median Median
Millimeters Millimeters Millimeters Millimeters Millimeters Millimeters Millimeters Millimeters Millimeters
Baseline Baseline n342 n342 n342 n338 n338 n338 n338 n338
Mean cIMT Mean cIMT 0.70?0.13 0.70?0.13 0.69 0.69?0.13 0.69?0.13 0.68 0.68 0.68 0.64
Mean maximum cIMT Mean maximum cIMT 0.80?0.16 0.80?0.16 0.78 0.80?0.17 0.80?0.17 0.76 0.76 0.76 0.94
24 months follow-up 24 months follow-up n320 n320 n320 n322 n322 n322 n322 n322
Mean cIMT Mean cIMT 0.70?0.14 0.70?0.14 0.69 0.71?0.15 0.71?0.15 0.71?0.15 0.68 0.68 0.29
Mean maximum cIMT Mean maximum cIMT 0.810.17 0.810.17 0.79 0.820.18 0.820.18 0.820.18 0.78 0.78 0.27
Difference baseline to 24 months Difference baseline to 24 months Difference baseline to 24 months Difference baseline to 24 months
Mean cIMT 0.0058?0.0037 0.0058?0.0037 0.0058?0.0037 0.0095 0.0111?0.0038 0.0111?0.0038 0.0111?0.0038 0.0058 0.0058 0.29
Mean maximum cIMT 0.0103?0.0049 0.0103?0.0049 0.0103?0.0049 0.0103 0.0175?0.0049 0.0175?0.0049 0.0175?0.0049 0.0160 0.0160 0.27
- Consistent inferential results observed for
non-parametric (median) and parametric (mean)
analyses - Exclusion of patients with missing
data or biologically implausible measures did not
change primary or secondary outcomes
8
Mean cIMT during 24 months of therapyLongitudinal
, repeated measures analysis
Mean IMT (mm)
ENHANCE
9
ENHANCE Safety

• Both regimens well tolerated, with overall safety
  profiles generally similar and consistent with
  product labels
• One case of viral hepatitis A in the
  simvastatin-only arm
• One case of myopathy (defined as CPK gt 10 ULN,
  with associated muscle symptoms) in the
  simvastatin-only arm and 2 cases in the
  Ezetimibe-Simvastatin arm

Simva 80 mg Simva 80/Eze 10 mg p
Consecutive n360 n356
ALT and/or AST 3 X ULN 8 10 0.62
CPK 10 X ULN 8 4 0.25
Subjects with 2 consecutive measurements for ALT
and/or AST a single last measurement 3 ULN a
measurement 3 X ULN followed by lt 2 ULN that
was taken more than 2 days after the last dose of
study medication
10
An Examination of the ENHANCE Trial
Evan A. Stein, MD, PhD Director, Metabolic and
Atherosclerosis Research CenterVoluntary
Professor of Pathology and Laboratory
MedicineUniversity of Cincinnati Cincinnati, OH
11
Questions after ENHANCE

• Why were the results of ENHANCE different from
  the earlier carotid IMT trials?

12
The Trial Design and Population
To have any chance using cIMT of demonstrating
that one treatment is better than another two
critical factors must be present

• The population studied must have significant and
  quantifiable lipid rich intima if minimal or no
  significant atherosclerosis is present then only
  possible change to be assessed is progression
• The control group must show significant
  progression if not then only significant
  regression in the test group can result in a
  positive trial

13
Critical Factors for Successful cIMT Trial
14
Critical Factors for Successful cIMT Trial
Simva progressed atorva regressed SUCCESS!!
ASAP - 1997
0.95
progression
progression
0.90
0.85
cIMT mm
0.80

• In ASAP the patient population had cIMT with
  mean baseline of 0.92 mm indicating presence of
  lipid enriched atherosclerosis
• In ASAP the control group of simvastatin
  showed significant progression of 0.036 mm/2yrs
  despite 40 reduction in LDLc

0.75
0.70
0.65
regression
regression
0
1
2
years
15
Critical Factors for Successful cIMT Trial
progression
Simva No progress Sim/Eze No regress FAILURE!!
regression
16
Critical Factors for Successful cIMT Trial
Identical FH populations 25 less
cIMT REGRESSION!!!
17
The Patient PopulationWhat changed from ASAP to
ENHANCE?

• In ASAP, the study upon which ENHANCE was based,
  the progression in cIMT over 2 years in the
  simvastatin 40 mg treated group (LDLc decrease
  40) was 0.036 mm over 2 years. At the end of the
  2 year ENHANCE trial the change in cIMT was
  substantially less than projected in both groups
  at 0.011 mm in the Eze/Simva group and 0.005 mm
  in the simvastatin 80 mg group (p0.29 NS).
• Thus a 40 reduction in LDLc both trials, done 5
  years apart, with the same drug, in the same
  patient population and similar entry LDLc levels
  resulted in 6 fold less progression of cIMT in
  ENHANCE
• The only difference in the simvastatin
  monotherapy groups between the two studies was
  the pre-treatment of patients and the amount of
  baseline atherosclerosis in ENHANCE.
• ASAP was recruited in 1997 in Holland from a
  large screening program for FH, only few statins
  (prava, simva, fluva) available in prior years
  and not aggressive Rx
• ENHANCE recruited from global well established
  lipid clinics with large and long identified and
  aggressively Rx patients

18
The Population
Thus in ENHANCE neither of the two critical
factors needed for any chance of a positive
outcome, for ezetimibe, or likely any other
treatment, existed!

• The population studied did not have significant
  and quantifiable lipid rich intima the majority
  would not even have qualified for any other IMT
  trial such as METEOR
• The control group did not show significant
  progression in fact the same treatment as in
  ASAP resulted in no progression over 2 years
• The most obvious and scientific explanation of
  ENHANCE results is that it was a failed trial and
  not a failed drug.
• If one were to test an antibiotic it is crucial
  to recruit patients with an infection, if not
  then no matter how good the antibiotic it will be
  impossible to show it is better than even
  placebo!

ENHANCE
19
The Patient PopulationWhat about the subgroup
with thick baseline cIMT?

• In neither the patients in the highest quintile
  or those above the median was there any
  difference does this invalidate the argument
  that the population had too little
  atherosclerosis?
• As there was no compositional studies of the
  intimal wall, only measurement of thickness, it
  is entirely possible, if not likely, that the
  amount of lipid within the wall even in these
  patients was minimal after many years of prior
  lipid lowering therapy and the thicker intima was
  due to non-lipid, more fibrous tissue, making it
  resistant to regression
• The data from the pre-specified analysis but
  using medians in the small number of subjects
  (20 of total) who had not received lipid
  lowering therapy prior to the trial, provides
  support for this as presumably their more lipid
  enriched intima showed a separation between the
  treatment groups in favor of more intense lipid
  and hs-CRP reduction.

20
No significant changes in those with no prior
statin Rx using means
Progression
Change cIMT (mm)
Regression
ENHANCE
21
Subgroup Analysis IMT Change by Reported Statin
Pretreatment medians
Median Change for Patients Reporting No Statin Use at Baseline Median Change for Patients Reporting No Statin Use at Baseline Median Change for Patients Reporting No Statin Use at Baseline
Median 25th , 75th percentiles

Simva 80 0.0033 -.0294, 0.0258
EZ/Simva 10/80 -0.0061 -.0299, 0.0357
Simva 80
EZ Simva 10/80
22
What about the trial indicating potential harm
from increased CVD events?

• Although ENHANCE was a relatively small trial in
  low risk FH patients was there any evidence from
  the CVD events that addition of ezetimibe caused
  harm?
• Can one even pick up such a signal from such
  small trials as ENHANCE in this FH population?

ENHANCE
23
Cardiovascular Events - ENHANCE
Incidence of cardiovascular events () Simvastatin Simvastatin plus Ezetimibe
(n363) (n357)
Cardiovascular death/Myocardial infarction/ Revascularization/Stroke 7 (1.9) 10 (2.8)
Cardiovascular death 1 (0.3) 2 (0.6)
Non-fatal myocardial infarction 2 (0.6) 3 (0.8)
Non-fatal stroke 1 (0.3) 1 (0.3)
Revascularization 5 (1.4) 6 (1.7)
24
CVD Events recent FH cIMT trialsCan these
small trials give some hint of CVD
harm?RADIANCE I (CETPi) and CAPTIVATE (ACATi)
Incidence of CVD events () Atorvastatin Atorva Torcet Statin Statin ACATi
(n454) (n450) (n438) (n443)
CVD death/MI/ Revasc/Stroke 11 (2.4) 23 (5.1) 15 (3.4) 28 (6.3)
RADIANCE I
CAPTIVATE
plt0.05
Thus even small studies (700-900 patients) in FH
appear to be able to detect potential CVD harm
Kastelein et al NEJM 2007 3561620-30
Meuwese MC et al AHA New Orleans 2007 (in press)
25
The CompoundEzetimibe lowers LDLc by a different
mechanism than statins?

• Reflects total ignorance of lipid physiology as
  the mechanism by which LDLc is reduced is in fact
  identical whether treated with statin (HMG CoA
  reductase inhibition), cholesterol absorption
  inhibition or bile acid sequestration!
• In all three modalities there is a depletion of
  hepatic intracellular cholesterol and the cell
  responds by upregulation of the LDL receptor
  doubtful if the carotid arterial wall can detect
  the mechanism by which receptor upregulation
  occurred!!
• LDLc reduction by other mechanisms, such as
  fibrates, niacin, MTP inhibition, Apo B antisense
  or CETP inhibition is different as none involve
  the LDL receptor.

ENHANCE
26
The CompoundEzetimibe lowers LDLc by a different
mechanism than statins?
Cholesterol
Hepatic cellular lipid regulation
27
The CompoundEzetimibe no pleiotropic effects?

• Pleotrophic effects are manifest through an
  impact on vascular inflammation and the best
  biomarker for inflammation and thus pleotrophism
  is accepted as hs-CRP
• In ENHANCE hs-CRP was reduced very robustly, in
  fact 50 more than the added LDLc reduction in
  the Eze/Simva group than simvastatin alone
• Effect on other apo B containing lipoproteins
  were robust and on HDLc was significantly
  (p0.05) increased by addition of ezetimibe

ENHANCE
28
Conclusion
The addition of Ezetimibe to Simvastatin did lead
to expected changes in LDL-c and hsCRP, but did
not reduce any cIMT parameter more than
simvastatin alone The most likely reason is that
the patient population had minimal lipid in their
carotid arteries and thus inability to show any
change no matter how much better the treatment of
one group was compared to the other There were
no safety differences between the groups and no
significant differences in cardiovascular events
ENHANCE
29
Conclusion
The most obvious and scientific explanation of
ENHANCE results is that it was a failed trial and
not a failed drug. While the results of ENHANCE
have been less than optimal for the sponsors of
the trial they actually carry very good news for
those with FH, and all patients on long term
lipid lowering therapy, in that the data would
seem to strongly indicate that even moderate,
long term LDLc lowering dramatically reduces the
atherosclerotic burden, at least in carotid
arteries, and virtually halts progression of the
underlying disease. Thus the glass is not half
empty but clearly more than half full!
ENHANCE
30
An Examination of the ENHANCE Trial
Michael H. Davidson, MD, FACC, FACP Clinical
Professor Director, Preventive Cardiology The
University of Chicago Executive Medical
Director Radiant Research Chicago, IL
31
Questions after ENHANCE

• Is there reason to suspect that ezetimibes
  mechanism of action is less effective than that
  of statins?

32
An Examination of the ENHANCE Trial
B. Greg Brown, MD, PhD Professor of Medicine and
Cardiology University of Washington School of
Medicine Seattle, WA
33
Questions after ENHANCE

• Can the ENHANCE results be interpreted in any way
  other than
• Ezetimibe is just an expensive placebo?

34
Comparisons of Carotid Plaque Tissue Composition
Two Groups of Eight Matched Patients


Triple Therapy x 10 years
Untreated
Total plaque area 58 mm
Total plaque area 64 mm
2
2
0.7 mm
2
10.2 mm
2
(1)
(17)
3 mm
2
p0.01
5 mm
2
(3)
(5)
6 mm
2
2 mm
2
(10)
(3)
46 mm
2
49 mm
2
(84)
(77)
2
Fibrous tissue area, mm2
( of total plaque)
Lipid plus calcium area, mm2
( of total plaque)
Lipid deposits area, mm2
( of total plaque)
Calcium cluster area, mm2
( of total plaque)
Zhao, X-Q, et al. ATVB 2001211623-29
35
An Examination of the ENHANCE Trial
Jennifer G. Robinson, MD, MPH Associate Professor
Departments of Epidemiology and Medicine
Director, Lipid Research Clinic University of
Iowa Iowa City, IA
36
Questions after ENHANCE

• Is there evidence to support the use of
  alternative LDL-lowering therapies?

37
Selection of patients for more aggressive
LDL-lowering Risk curve concept
Robinson JG, Stone NJ. Am J Cardiol 2006 98
1405-1408
38
Add-on to statin therapy Drug options to ?LDL
?Non-HDL-C
Drug LDL-C Non-HDL-C Trigs HDL-C
Double statin dose -6 -6 -2 to -12 -2 to 2
Ezetimibe 10 mg -15 -12 -9 NS
Niacin 2 gr -14 -31 -24 18
Bile acid binding agent Colestipol 2 scoops (6 gr) Cholestyramine 2 scoops (8 gr) Coleselvalam 6 tabs (3.75 gr) -12 -5-8 0 to 23 1-7
Fenofibrate 145 mg -6 to 4 -3 to -18 -15 to -20 13
Gemfibrozil 600 mg BID 7 2 -18 0
Jones PH, et al. Am J Cardiol 2003 92
152-60. Robinson JG, Davidson MH. Expert Rev
Cardiovasc Ther. 2006 4 461-76 Kos Niaspan
prescribing information 2005 Sankyo Welchol
prescribing information 2005 Athyros VG et al.
Diabetes Care 2002 25 1198-1202 Durrington PN
et al. Diab Res Clin Pract 2004 64
137-51. Wagner AM, et al. J Clin Endocrinol Metab
2003 88 3212-17. Estimated Total
cholesterol-HDL
39
Consistent relationship betweenLDL-C reduction
and CHD relative risk for all LDL-lowering
treatments
Cholestyramine Colestipol
Robinson JG et al. J Am Coll Cardiol.
20054618551862.
40
Niacin RCTs
Unpublished
41
PPAR agonists Expected CVD risk reduction from
LDL HDL changes
Robinson JG. PPAR Review 2008 in press
42
Gemfibrozil vs Fenofibrate Adverse Events
(excluding reports with concomitant cerivastatin
use) submitted to the US Food and Drug
Administration from January 2000 to December 2004
Adverse Event Reports Gemfibrozil Gemfibrozil Fenofibrate Gemfib vs Feno OR 95 CI P-value
Rates/million prescriptions Rates/million prescriptions Rates/million prescriptions
Gemfibrozil better
All 31.0 40.0 40.0 0.76 0.69  0.83 lt0.001
Serious 20.0 27.9 27.9 0.72 0.65  0.81 lt0.001
Liver 2.6 6.9 6.9 0.37 0.28  0.50 lt0.001
Fenofibrate better
Rhabdomyolysis 9.7 3.6 3.6 2.67 2.11  3.39 lt0.001
Muscle-related with no rhabdomyolysis 8.1 5.8 5.8 1.36 1.12  1.71 0.002
Holoshitz N, et al. Am J Cardiol 2008 101 95-97
43
An Examination of the ENHANCE Trial
Jerome Cohen, MD, FACC, FAHA, FACP Chair, NLA
Consumer Affairs Committee Professor Emeritus St.
Louis University School of Medicine St. Louis, MO
Hosted by the Midwest Lipid Association
44
Questions after ENHANCE

• How should ezetimibe continue to be used
• in clinical practice?