Clinical Therapeutics/Volume 30, Number 12, 2008

Comparative Effects of lO-mg Versus 80-mg Atorvastatin on

High-Sensitivity C-Reactive Protein in Patients with Stable
Coronary Artery Disease: Results of the CAP (Comparative
Atorvastatin Pleiotropic Effects) Study
Jacques Bonnet, MD1; R. McPherson, MD, PhD2; A. Tedgui, MD3; D. Simoneau, MD4;
A. Nozza, MSc5; P. Martineau, PharmD5; and Jean Davignon, MD6; forthe CAP Investigators*
1H6pital du Haut-Leveque, Pessac, France; 2University of Ottawa Heart Institute, Ottawa, Ontario, Canada;
31NSERM, Paris, France; 4Medical Division, Pfizer France, Paris, France; sMedical Division, Pfizer Canada,
Kirkland, Quebec, Canada; and 6C!inical Research Institute of Montreal, Montreal, Quebec, Canada

ABSTRACT time points, were also calculated. The secondary effi-

Background: The major beneficial effect of statins-
reducing the risk for coronary events-has primar-
ily been ascribed to reductions in low-density
lipoprotein cholesterol (LDL-C) but may in part be
related to a direct antiinflammatory action (ie, de-
creased high-sensitivity C-reactive protein [hs-CRP]
concentration).
Objectives: The objectives of this CAP (Comparative
Atorvastatin Pleiotropic Effects) study were to com-
pare the effects of low- versus high-dose atorvastatin
on hs-CRP concentrations and to determine the rela-
tionship between changes in LDL-C and hs-CRP con-
centrations in patients with coronary artery disease
(CAD), low-grade inflammation, and normal lipo-
protein concentrations.
Methods: This multicenter, prospective, randomized,
double-blind, double-dummy study was conducted at
65 centers across Canada and Europe. Patients with
documented CAD, low-grade inflammation (hs-CRP
concentration, 1.5-15.0 mg/L), and a normal-range
lipid profile (LDL-C concentration, 1.29-3.87 mmol/L
[50-150 mg/dL]; triglyceride [TG] concentration,
<4.56 mmoUL [<400 mg/dL]) were randomly assigned
to receive 26-week double-blind treatment with ator-
vastatin 10 or 80 mg QD. Investigators were to aim for
the National Cholesterol Education Program Adult
Treatment Panel III (NCEP ATP III) LDL-C target of
<2.59 mmol/L (<100 mg/dL). The primary end point
was the percentage change from baseline in hs-CRP,
as measured at baseline and weeks 5, 13, and 26 using
high-sensitivity, latex microparticle-enhanced immu-
noturbidimetric assay. Changes from baseline in
LDL-C, as measured directly in serum at the same
cacy variables included the percentage changes from
baseline in lipid parameters (LDL-C, high-density lipo-
protein cholesterol [HDL-C], total cholesterol [TC], TG,
apolipoprotein B, non-HDL-C, and TC:HDL-C ratio)
at 5, 13, and 26 weeks of treatment. Tolerability was
assessed using physical examination, including vital
sign measurement, and laboratory analyses.
Results: A total of 339 patients were enrolled
(283 men, 56 women; mean age, 62.5 years; weight,
81.3 kg; 10-mg/d group, 170 patients; 80-mg/d group,
169). No significant differences in baseline demograph-
ic or clinical data were found between the 2 treatment
arms. In the 10-mg group, hs-CRP was decreased
by 25.0% at 5 weeks and remained stable thereafter
(%1'1 at week 26, -24.3%; P < 0.01). In the 80-mg
group, hs-CRP was decreased by 36.4% at 5 weeks and
continued to be decreased over the study period
(%1'1, -57.1 % at week 26; P < 0.001 vs baseline). At
5 weeks, LDL-C was decreased by 35.9% in the
10-mg group and by 52.7% in the 80-mg group
(P < 0.001 between groups) and remained stable
thereafter (%1'1 at week 26, -34.8% and -51.3%, re-
spectively; P < 0.001 between groups). The NCEP
ATP III LDL-C target of <2.59 mmol/L (<100 mg/dL)
was reached in 77.1 % of patients treated with atorva-
statin 10 mg and 92.3% of those treated with 80 mg
(P < 0.001). Dual targets of hs-CRP <2 mg/L and
*The CAP Investigators are listed In the Acknowledgments.
Accepted for publicatIOn October 17, 2008
dOl:1 0.1 016/J.cIInthera.2008.12.023
0149-2918/$32.00
© 2008 Excerpta Medica Inc. All nghts reserved.

2298 Volume 30 Number 12


LDL-C <1.81 mmol/L «70 mg/dL) were reached in
a significantly greater proportion of patients in the
80-mg group compared with the 10-mg group (55.6%
vs 13.5%; P < 0.001). The decrease in hs-CRP was
largely independent of baseline LDL-C and change
in LDL-C. Two serious adverse events were reported
by the investigator as treatment related: acute hepati-
tis in the 10-mg group and intrahepatic cholestasis in
the 80-mg group, in 2 patients with multiple comor-
bidities. Two deaths occurred during the study, both in
the atorvastatin 80-mg group (1, myocardial infarc-
tion; 1, sudden death), neither of which was deemed
treatment related by the investigator.
Conclusions: In these patients with documented
CAD, evidence of low-grade inflammation, and normal-
range lipid profiles, the effects of atorvastatin on chang-
es in hs-CRP were dose dependent, with the high dose
(80 mg) being associated with significantly greater re-
ductions in hs-CRP concentrations. Both doses were
associated with a significant and progressive decline
in hs-CRP largely independent of changes in LDL-C,
HDL-C, and TG. Clinical Trials Identification Number:
NCT00163202. (Clin Ther. 2008:30:2298-2313) ©
2008 Excerpta Medica Inc.
Key words: cardiovascular diseases, cholesterol, lipids,
lipoproteins, C-reactive protein.
INTRODUCTION
In individuals with and without established coronary ar-
tery disease (CAD), C-reactive protein (CRP) is a robust
marker of inflammation, and plasma high-sensitivity
(hs)-CRP has been found to be a predictor of risk for
coronary events. 1 Evidence suggests that CRP may be
involved in the atherothrombotic process. I - 4 CRP is
also believed to be a regulator of endothelial function,
vascular remodeling, and thrombosis. 4- 7
Statin treatment has been associated with reduced
plasma cholesterol concentrations; in a meta-analysis
of 14 trials in 90,056 patients, 8 statins were associated
with a mean low-density lipoprotein cholesterol (LDL-C)
difference at 1 year of 1.09 mmol/L compared with
controls. In AFCAPSlTexCAPS (Air ForcelTexas Coro-
nary Atherosclerosis Prevention Study),9 5742 pa-
tients received lovastatin 20 to 40 mg/d or placebo for
a mean duration of 5.2 years. A significantly greater
reduction in the relative risk for coronary events was
found in patients with low (less than the median)
LDL-C concentration who had a high (greater than
December 2008
J. Bonnet et al.
the median) hs-CRP concentration (0.58 mg/L [95%
CI, 0.34-0.98 mg/L]) compared with those with low
(less than the median) hs-CRP (1.08 mg/L [95% CI,
0.56-2.08 mg/L ]).10
Results from AFCAPSlTexCAPSlO and the CARE
(Cholesterol and Recurrent Events) triaP1,12 have sug-
gested that the decreases in CRP with statin use may
be independent of LDL-C concentrations. In AFCAPS/
TexCAPS, the effect of lovastatin on hs-CRP concen-
tration was not found to be significantly related to the
effect of lovastatin on lipid levels; the Spearman cor-
relation coefficients for the relationship between the
percentage change in hs-CRP level and the percentage
change in total cholesterol (TC) and LDL-C concen-
trations were -0.001 and 0.014, respectively. 10 In the
CARE trial,11,12 at 5 years of follow-up, no significant
correlation was observed between the magnitude of
change in hs-CRP and the magnitude of change in
LDL-C among 472 patients who received pravastatin
40 mg/d or placebo. 12
In 4162 patients with acute coronary syndrome
(ACS) enrolled in the PROVE IT-TIMI 22 (Pravastatin
or Atorvastatin Evaluation and Infection Therapy-
Thrombolysis in Myocardial Infarction 22) trial,13 those
with hs-CRP concentrations <2 mg/L after treatment with
pravastatin 40 mg/d or atorvastatin 80 mg/d for a
mean duration of 24 months had significantly fewer
recurrent events compared with those with hs-
CRP concentrations >2 mg/L (2.8 vs 3.9 events per
100 person-years; P = 0.006), regardless of the LDL-C
concentration reached. 14 Reductions in hs-CRP were
significant at 1 month in PROVE IT-TIMI 22 but
did not reach significance until 4 months in the A
to Z trial 15 in 4497 patients who received simvastat-
in (40 mg/d for 1 month, then 80 mg/d for 23 months)
or 4-month placebo followed by 20 mg/d for the re-
mainder of the trial. 16,17 At the same time, there was
an early (0-4 months) favorable separation of event
curves in PROVE IT-TIMI 22 (hazard ratio [HR],
0.80), but not in A to Z (HR, 1.17), suggesting a pos-
sible benefit associated with the early CRP reduc-
tion. 16 ,17 In the MIRACL (Myocardial Ischemia Re-
duction with Aggressive Cholesterol Lowering) study, 18
16-week administration of high-dose atorvastatin
(80 mg/d) versus placebo was associated with a signifi-
cant reduction in CRP (-86% vs -78%; P < 0.001) in
patients with ACS and LDL-C <3.2 mmol/L.1 9
Although the results from those previously pub-
lished studies 10,12,14,16,17,19 have suggested that statin
2299
 Clinical Therapeutics
use is associated with decreased hs-CRP concen-
trations, a randomized controlled trial (RCT) in
107 patients with stable CAD and high (>3 mg/L) hs-
CRP levels did not find a dose-response relationship
between simvastatin 80 mg or 20 mg/d for 12 weeks
and decreases in hs-CRP levels (-0.6 vs -0.5 mg/L).20
A meta-analysis of 14 randomized trials in 90,056 pa-
tients found that statin treatment was associated with
a significantly reduced incidence of major coronary events
compared with controls (21 %; rate ratio = 0.79; 95%
CI, 0.77-0.81; P < 0.001).8 The clinical benefit of stat-
ins may be related not only to their lipid-lowering ef-
fects but also to other pleiotropic effects, such as anti-
inflammatory, immunomodulatory, antithrombotic,
and vascular effects. 21
The objectives of this Comparative Atorvastatin
Pleiotropic Effects (CAP) study were to compare the
effects of low- versus high-dose atorvastatin on hs-CRP
concentrations, and to determine the relationship be-
tween changes in LDL-C and hs-CRP concentrations in
patients with stable CAD, evidence of low-grade in-
flammation, and normal lipoprotein concentrations.
PATIENTS AND METHODS
Study Design
This CAP study used a 26-week, multicenter, pro-
spective, randomized, double-blind, double-dummy
design and was conducted at 65 sites in Canada and
Europe. The relevant institutional review boards ap-
proved the protocol, and written informed consent
was obtained from all patients before study entry. This
study was conducted in compliance with the ethical
principles of the Declaration of Helsinki. 22
Inclusion/Exclusion Criteria
Eligible patients included men and women aged
<80 years with documented CAD, low-grade inflam-
mation (hs-CRP concentration, 1.5-15.0 mg/L), and a
normal-range lipid profile (LDL-C concentration, 1.29-
3.87 mmol/L [50-150 mg/dL]; triglyceride [TG] con-
centration, :c:;4.56 mmol/L [:c:;400 mg/dL]). CAD was
defined as the presence of ~1 of the following: history
of myocardial infarction (~3 months before screening),
stable angina, history of unstable angina, history of
percutaneous transluminal coronary angioplasty and!
or coronary artery bypass grafting (~6 months before
screening), and/or coronary stenosis ~50%.
Patients were considered ineligible for the study if
they met any of the following criteria: presence of
2300
secondary hyperlipidemia or type 1 diabetes mellitus
(DM) or type 2 DM with insulin therapy, inadequately
controlled DM (glycosylated hemoglobin, >9%), body
mass index >32 kg/m 2, history or presence of alcohol
or drug abuse, inability to comply with study proce-
dures, progressive or life-threatening disease with a
life expectancy of <1 year, contraindication to statin
treatment (history of intolerance or hypersensitivity
to statins, active hepatic disease, or hepatic dysfunc-
tion), and/or current treatment with a potent cyto-
chrome P450 3A4 inhibitor (eg, erythromycin, keto-
conazole). Women who were pregnant or breastfeeding
were excluded. Eligible women of childbearing poten-
tial were required to use an effective method of con-
traception throughout the study period.
Study Drugs
After an initial 6-week washout period in patients
who had been receiving statin treatment, fibrates, nia-
cin, and/or resins prior to selection, patients were
randomized in a 1:1 ratio to receive one of the follow-
ing treatment schedules: two 40-mg atorvastatin tab-
lets and 1 placebo tablet of atorvastatin 10 mg, or one
10-mg atorvastatin tablet and 2 placebo tablets of
atorvastatin 40 mg (Figure 1). Patients were random-
ized at visit 2, with treatment being allocated in nu-
meric order. The double-blind study drug supplies
were delivered to investigators in sealed, numbered
randomization envelopes according to the computer-
generated randomization code.
Patients could continue, at a stable dosing regimen,
their usual antihypertensive and antianginal treatment
during the study. Patients could receive short (:C:;7 days)
courses of antiinflammatory drugs (NSAIDs, cyclo-
oxygenase-2 inhibitors, or corticosteroids) if treat-
ment was discontinued ~1 week before each hs-CRP
assessment (weeks 0 [baseline], 5, 13, and 26). Long-
course (~7-day) treatment with these agents was not
permitted. Treatment with acetylsalicylic acid at doses
of :c:;325 mg/d was permitted.
Efficacy Assessments
The primary efficacy end point was the percentage
change from baseline to 26 weeks in hs-CRP concen-
trations with atorvastatin 10 and 80 mg/d. The sec-
ondary efficacy variables included the percentage
changes from baseline in lipid parameters (LDL-C,
high-density lipoprotein cholesterol [HDL-C], TC, TG,
apolipoproteinB [apoB],non-HDL-C,andTC:HDL-C
Volume 30 Number 12
 J. Bonnet et al.

Randomized to treatment
(N ~ 340)

Atorvastatm 10 mg Atorvastatm 80 mg
(n~170) (n ~ 169*)

Completed Withdrawn (n ~ 13) Completed Withdrawn (n ~ 16)

(n~157) Adverse event t (7) (n ~ 153) Adverse event§ (10)
Protocol violation (2) Withdrawn consent (4)
Withdrawn consent (2) Death (2)
Other* (2)

Figure 1. Study design and disposition of patients assigned to receive 26-week treatment with atorvastatin
10 or 80 mg in patients with stable coronary artery disease (modified intent-to-treat [ITT] popula-
tion). *One patient was not included in the ITT analysis due to lack of postbaseline data. tlncluded
asthenia and myalgia (2 patients); cerebrovascular accident (2 patients); and atrial fibrillation and
palpitations, chest pain, and leg cramps (1 patient). :j:lncluded loss to follow-up and investigator
decision based on patient noncompliance (1 patient each). §Included myalgia (2 patients), and
weight loss, asthenia, abdominal pain, anemia, liver damage, dyspepsia, increased alanine amino-
transferase, atrial fibrillation, and myalgia (1 patient).

ratio) at 5, 13, and 26 weeks of treatment. The rela- Tolerability Assessment

tionship between changes in hs-CRP and LDL-C was
also evaluated.
Laboratory Analyses
For hs-CRP assessment, blood samples were drawn
by a registered nurse or phlebotomist as per usual prac-
tice at each site into dry tubes and EDTA tubes. The
tubes were centrifuged for 10 minutes at 1500g and
stored at 4 D C or -20 D C, depending on which test was
being run, until shipped on dry ice to a centrallabora-
tory (Focus Bio-Inova Global Laboratory Services,
Herndon, Virginia). hs-CRP was measured with high-
sensitivity, latex microparticle-enhanced immunotur-
bidimetric assay using a commercial kit (Tina-quant,
Roche Diagnostics GmbH, Mannheim, Germany).
TC and TG were measured using an enzymat-
ic colorimetric assay. LDL-C and HDL-C were mea-
sured directly using a homogeneous enzymatic colo-
rimetric assay, and apoB was measured using an
immunoturbidimetric assay (Roche Diagnostics
GmbH).
Tolerability analyses included adverse events re-
ported and/or observed during clinical examination,
which included monitoring of vital signs (heart rate,
blood pressure, electrocardiography and laboratory
analyses (tolerability parameters, including alanine
aminotransferase [ALT] , aspartate aminotransferase
[AST], and creatine kinase [CK]).
Statistical Analysis
Primary efficacy analysis was performed on the
modified intent-to-treat (mITT) population, which con-
sisted of all patients who were randomized to treat-
ment, received ~1 dose of assigned treatment, and had
data available from ~1 postbaseline assessment of the
primary end point (hs-CRP concentration). Missing
values were replaced according to the last-observation-
carried-forward (LOCF) approach. Patients with no
on-treatment data for a parameter were excluded from
the analysis of that parameter. The per-protocol popu-
lation consisted of all patients randomized to treatment
who received study drug ~20 weeks and had no major

December 2008 2301

 Clinical Therapeutics
protocol deviations. The tolerability analysis included
all patients who were randomized to treatment and
received ~1 dose of assigned treatment.
The hs-CRP concentrations were not normally dis-
tributed and therefore are reported as geometric means.
Percentage changes from baseline (mean of screening
and baseline values) to week 26 were compared be-
tween the 2 treatment groups using a 2-tailed Mann-
Whitney U test. 23 ,24 Changes from baseline in lipid
parameters were assessed using the t test.
Spearman's rank correlation coefficient23 ,24 was
used to test for correlations between changes in hs-
CRP and LDL-C. Significance level was fixed at 0.05.
Because this was an exploratory study, no adjustment
for multiplicity was used.
A post hoc analysis was conducted to determine
the proportions of patients in whom the most recently
proposed National Cholesterol Education Program
(NCEP) targets (hs-CRP, <2 mg/I)4; LDL-C, <100 mg!
dL [<70 mg!dLj25,26) were reached.
RESULTS
Patient Disposition
Between May 2002 and January 2005,1198 patients
were screened and 340 were randomized to receive
treatment at 65 sites in France, Canada, Poland, the
Czech Republic, Romania, Slovakia, and Russia. Of the
858 screening failures, 806 did not meet entry criteria
and 37 withdrew consent. Consequently, the tolerability
population included 340 patients, and the mITT popu-
lation included 339. The proportions of patients who
completed the study were 92.4% in the 10-mg group
and 90.5% in the 80-mg group (Figure 1).
There were no meaningful significant differences
between the 2 treatment groups in baseline demo-
graphic or clinical characteristics (Table I).
Primary End Point
The mean (95% CI) baseline hs-CRP concentra-
tion was 3.1 (2.8-3.5) mg/L and 3.6 (3.2-3.9) mg/L
in the atorvastatin 10- and 80-mg groups, respec-
tively (Figure 2). In the 10-mg group, the mean per-
centage change in hs-CRP at 5 weeks was -25.0%
(-32.7% to -16.6%); hs-CRP remained stable there-
after. In the 80-mg group, hs-CRP changed by
-36.4% (-43.8% to -28.0%) at 5 weeks and contin-
ued to decrease for the duration of the study period,
with a total change of -57.1 % (-62.3% to -51.2%)
at 26 weeks (P < 0.001 vs baseline). At study end
2302
(LOCF), the mean changes in hs-CRP were -20.8%
(-31.6% to -8.3 %) in the 10-mg group and -54.9%
(-60.6% to -48.5%) in the 80-mg group.
Secondary End Points
The effects of atorvastatin on plasma lipid concen-
trations and apoB are outlined in Figure 3. At study
end, the mean (95% CI) percentage changes in LDL-C
were -34.8% (-37.5% to -32.0%) with the 10-mg
dose and -51.3% (-54.5% to -48.1 %) with the 80-mg
dose. These decreases were evident at the first visit
(5 weeks) and remained stable thereafter (Figure 3).
TC, TG, apoB, non-HDL-C, and TC:HDL-C ratio
were all found to have a similar pattern, decreasing by
5 weeks and remaining stable until study end, with
changes from baseline being significantly greater with
atorvastatin 80 mg compared with 10 mg for each of
these lipid parameters at all time points (P < 0.005).
As expected, changes in HDL-C were less consistent,
with a mean change at study end of +3.6% with the
10-mg dose and +0.8% with the 80-mg dose (P = 0.01).
On analysis according to tertiles of baseline hs-
CRP concentrations, those in the highest tertile were
found to have the greatest changes from baseline in
hs-CRP (-37.2% and -69.3% in the 10- and 80-mg
groups, respectively; P < 0.001) (Figure 4).
Correlation Between Reduction in hs-CRP and
Lipid Parameters
The decrease in hs-CRP was largely independent
of baseline LDL-C and change in LDL-C and totally
independent of changes in HDL-C and TG, regardless
of dose (Table II). Although the P value suggested
statistical significance, the correlation between hs-
CRP and LDL-C was small, with <5% of change in
hs-CRP being explained by changes in LDL-C. In the
10-mg group, hs-CRP concentrations were significantly
correlated with LDL-C concentrations (r2 = 0.032;
P = 0.016), whereas the change in hs-CRP was not
correlated with change in LDL-C (r2 = 0.0036; P = NS)
(Table II). In the 80-mg group, the hs-CRP and LDL-C
concentrations (r2 = 0.0484; P = 0.005) and changes
in hs-CRP and LDL-C were significantly correlated
(r2 = 0.0484; P = 0.003).
LDL-C and hs-CRP Treatment Targets
The proportions of patients in whom recommend-
ed targets for LDL-C were reached are shown in
Figure 5. The NCEP Adult Treatment Panel (ATP) III
Volume 30 Number 12
 J. Bonnet et al.

Table I. Baseline demographic and clinical characteristics of the study population.

Atorvastatin 10 mg Atorvastatin 80 mg
Characteristic (n = 170) (n = 169)

Age, mean (SO), y 62.8 (8.9) 62.3 (9.7)

Sex, no. (%)
Male 144 (84.7) 139 (82.2)
Female 26 (15.3) 30 (17.8)
Weight, kg
Mean (SO) 81.5 (12.3) 80.8 (12.6)
Range 51.0-111.0 49.0-110.0
BMI, kgjm 2
Mean (SO) 27.8 (3.3) 27.8 (3.1)
Median (range) 27.7 (27.2-28.2) 27.7 (27.3-28.2)
Vital signs, mean (95% CI)
OBP, mm Hg 77.1 (75.9-78.3) 77.3 (76.1-78.5)
SBP, mm Hg 129.4 (127.3-131.5) 128.9 (126.5-131.4)
Heart rate, bpm* 63.6 (62.1-65.1) 65.8 (64.3-67.3)
Lipid parameters, mean (95% CI)
LOL-C
mmoljL 3.26 (3.16-3.34) 3.26 (3.19-3.34)
mgjdL 126 (122-129) 126 (123-129)
HOL-C
mmoljL 1.24 (1.22-1.30) 1.24 (1.19-1.30)
mgjdL 48 (47-50) 48 (46-50)
Non-HOL-C
mmoljL 4.16 (4.06-4.27) 4.10 (3.99-4.20)
mgjdL 160.62 (156.76-164.86) 158.30 (154.05-162.16)
TCjHOL-C ratio 4.59 (4.40-4.77) 4.58 (4.40-4.75)
ApoBt
mmoljL 5.52 (5.39-5.65) 5.39 (5.26-5.49)
mgjdL 213 (208-218) 208 (203-212)
TC
mmoljL 5.41 (5.31-5.52) 5.34 (5.23-5.44)
mgjdL 209 (205-213) 206 (202-210)
TG
mmoljL 2.05 (1.89-2.20) 1.85 (1.72-1.98)
mgjdL 181 (167-195) 164 (152-175)
Inflammatory marker, GM (95% CI)
hs-CRP, mgjL 3.13 (2.84-3.45) 3.56 (3.22-3.94)
Smokers, no. (%) 25 (14.7) 20 (11.8)

BMI ~ body mass Index; DBP ~ diastolic blood pressure; SBP ~ systolic blood pressure; LDL-C ~ low-density lipoprotein
cholesterol; HDL-C ~ high-density lipoprotein cholesterol; TC ~ total cholesterol; ApoB ~ apollpoproteln B; TG ~ tnglyc-
endes; GM ~ geometnc mean; hs-CRP ~ high-sensitivity C-reactive protein.
*P ~ 0.024 (Mann-Whitney U test).
t n ~ 161 (10 mg) and n ~ 160 (80 mg).

December 2008 2303

 Clinical Therapeutics

- . Atorvastatln 10 mg (n ~ 170)
• Atorvastatln 80 mg (n ~ 169)
A B
4
Study Week
2 4 6 8 10 12 14 16 18 20 22 24 26 LOCF
___-'------J-----'-------"-----'-------"------'------J-----'-------"-----'-------"------'--,{Ll
:=J 3
~
E
0...
• 0...
-10

0::: 2 0::: t
l( -20
l(
Vl
..c • Vl
..c
t t •

~ ~ -30
Q Q
c
-<I -40
?f2.
o ~
-50
o 2 4 6 8 10 12 14 16 18 20 22 24 26 LOCF

Study Week -60

•
Figure 2. (A) Geometric mean (GM) high-sensitivity C-reactive protein (hs-CRP) concentrations over 26-week
treatment with atorvastatin 10 or 80 mg in patients with stable coronary artery disease. P = 0.001
(last observation carried forward [LOCF]). (B) Percentage changes (%L'.) from baseline in GM hs-
CRP over time (modified intent-to-treat population; LOCF). *p = 0.008 versus 80 mg; tp < 0.001
versus 80 mg (2-tailed Mann-Whitney U test). P values are for change or %Il from baseline between
groups.

LDL-C target of <2.59 mmol/L (<100 mg/dL)25 was
reached in 92.3% (95% CI, 87.2%-95.8%) of patients
who received atorvastatin 80 mg and 77.1 % (95% CI,
70.0%-83.1 %) of those who received the 10-mg dose
(P < 0.001). The more stringent NCEP ATP III target of
<1.81 mmol/L «70 mg/dL)26 was reached in 81.1%
(95% CI, 74.3%-86.7%) of patients with the 80-mg
dose and 33.5% (95% CI, 26.5%-41.2%) with the
10-mg dose (P < 0.001). The percentage of patients in
whom an hs-CRP <2 mgIL14 was reached was signifi-
cantly higher in the 80-mg group than in the 10-mg
group (63.9% [95% CI, 56.2%-71.1%] vs 47.1%
[95% CI, 39.4%-54.8%]; P = 0.002).
Overall, dual targets of hs-CRP <2 mg/L and LDL-C
<1.81 mmol/L «70 mg/dL) were reached in 13.5%
(95% CI, 8.8%-19.6%) and 55.6% (95% CI,47.8%-
63.3%) of patients in the 10- and 80-mg groups, re-
spectively (P < 0.001).
Tolerability
The prevalences of treatment-related adverse events
were 8.2% in the 10-mg group and 11.8% in the 80-mg
group (Table III). The most common all-causality events
reported, with a global incidence >3 % in either the
10- or 80-mg group, were chest pain (4.1 % and 3.5%,
respectively), flulike symptoms (3.5% and 1.8%), ar-
thralgia (3.5% and 1.2%), respiratory tract infection
(2.9% and 5.9%), myalgia (1.8% and 3.5%), vagini-
tis (0% and 3.3%), and vulvovaginal disorder (0%
and 3.3 %). The most common treatment-related
events reported, with a global incidence> 1% in either
the 10- or 80-mg group, were asthenia (1.8 % and
1.8%, respectively), increased CK (1.8% and 0.6%),
myalgia (1.2% and 2.4 %), constipation (1.2% and
1.8%), increased AST (0% and 1.2%), and insomnia
(0% and 1.2%). The majority of events were of mild
to moderate intensity, with severe adverse events re-
ported in 1.2% of patients in both groups (1 patient
each: myalgia, dyspepsia, liver damage, and increased
CK). Overall, 2.6% of patients discontinued atorva-
statin use due to treatment-related adverse events
(5 patients: asthenia and/or myalgia; 1 patient each:
leg cramps, liver damage, dyspepsia, increased ALT).
Treatment-related adverse events involving the mus-

2304 Volume 30 Number 12

 J. Bonnet et al.

___ Atorvastatln 10 mg (n ~ 170)

A B • Atorvastatln 80 mg (n ~ 169)
4 1.5

••
3
::J' ::J'
........ ........ 1.0
0 0
E E
~
E 2 • ~
E
U
~
0
--.J
• U
~
0 0.5
I

o h o h
o 2 4 6 8 10 12 14 16 18 20 22 24 26 LOCF o 2 4 6 8 10 12 14 16 18 20 22 24 26 LOCF

Study Week Study Week

C D
6 3

::J'
........
4 • ::J' 2
........
0
E 3
• 0
E •
E E
~

~ 2
~

lJ
f-
•

o h o h
o 2 4 6 8 10 12 14 16 18 20 22 24 26 LOCF o 2 4 6 8 10 12 14 16 18 20 22 24 26 LOCF

Study Week Study Week

Figure 3. Mean lipid concentrations over 26-week treatment with atorvastatin 10 or 80 mg in patients with
stable coronary artery disease. (A) Low-density lipoprotein cholesterol (LDL-C) (P < 0.001).
(B) High-density lipoprotein cholesterol (HDL-C) (P = 0.01). (C) Total cholesterol (TC) (P < 0.001).
(D) Triglycerides (TG) (P < 0.001). P values are for percentage change from baseline between the
groups (last observation carried forward [LOCF]) and were calculated using the t test.
(continued)

culoskeletal system occurred in 3.5% of patients. The
most common adverse event involving the musculo-
skeletal system was myalgia (1.2 % and 2.4 % with the
10- and SO-mg doses, respectively).
The prevalences of any increase in AST/ALT >3 x up-
per limit of normal (ULN) from a normal baseline value,
regardless of causality, were 0% and 1.S% in the 10- and
SO-mg groups, respectively (Table III). There was 1 case of
CK >10 x ULN in each treatment group (0.6% overall).
Two serious adverse events were reported by the
investigator as treatment related: acute hepatitis in the
10-mg group and intrahepatic cholestasis in the SO-mg
group, in 2 patients with multiple comorbidities. Two
deaths occurred during the study, both in the atorva-

December 2008 2305

 Clinical Therapeutics

___ Atorvastatln 10 mg (n ~ 170)

E F • Atorvastatln 80 mg (n ~ 169)
6 6

5 5
::J'
4 • "-
0
E
E
4
::J'
"-
0
3
• ~

l{ 3
E
E •
l:fY
0...
<{
2
--.J
0
:::n:
2 •
Z
0 h 0 h
0 2 4 6 8 10 12 14 16 1820222426 LOCF 0 2 4 6 8 10121416 1820222426 LOCF

Study Week Study Week

G
6

0
.;:;
ro 4
0:::
U
3
•
~
0
I
•
U 2
f-

o h
o 2 4 6 8 10 12 14 16 18 20 22 24 26 LOCF

Study Week

Figure 3 (continued). Mean lipid concentrations over 26-week treatment with atorvastatin 10 or 80 mg in pa-
tients with stable coronary artery disease. (E) Apolipoprotein B (apoB) (P < 0.001). (F) Non-HDL-C
(P < 0.001). (G) TC/HDL-C ratio (P < 0.001) (modified intent-to-treat population [LOCF]). Pvalues are
for percentage change from baseline between the groups (LOCF) and were calculated using the t test.

statin SO-mg group (1, myocardial infarction; 1, sud-
den death), neither of which was deemed treatment
related by the investigator.
DISCUSSION
Plasma hs-CRP concentration is a biochemical marker
of inflammation in patients with established atheroscle-
rosisY It has been suggested that CRP is not only a
prognostic indicator of CAD risk but also may be caus-
ally involved in the atherothrombotic process and hence
could be considered a direct therapeutic target. 4 CRP
can be detected within atherosclerotic lesions,28 and al-
though not apparently directly atherogenic in rodent
models,4 CRP may have prothrombotic effects. 29 ,30

2306 Volume 30 Number 12

 J. Bonnet et al.

D Atorvastatln 10 mg (n - 170)
D Atorvastatl n 80 mg (n - 169)

1st Tertde 2nd Tertde 3rd Tertde

30

10
I I I

CL
0:::
-10 L - L -

l(
Vl
...c -30
c * t - -
- -
-<I
Cf2, -
-50 '-----

-70

-90
+
Figure 4. Percentage changes (%L'.) from baseline (95% CI) in geometric mean high-sensitivity C-reactive pro-
tein (hs-CRP) concentrations over 26-week treatment with atorvastatin 10 or 80 mg in patients with
stable coronary artery disease, by baseline hs-CRP tertile (modified intent-to-treat population, last
observation carried forward). First tertile = hs-CRP <2.3 mg/L; 2nd tertile = 2.3-3.9 mg/L; 3rd ter-
tile = >3.9 mg/L. *p < 0.001 versus 80 mg; tp < 0.015 versus 80 mg.

However, recent studies have not found a relation-
ship between genetic variants related to hs-CRP con-
centrations and CAD risk. 31,32 An analysis of associa-
tions between 3 genetic polymorphisms (-717C>T,
rs2794521; +1059G>C, rs1800947; +1444C>T,
rsl130864) at the CRP locus and related haplotypes
in 3252 patients with ACS31 found that while some
genetic variants were associated with circulating CRP,
none were consistently associated with the prevalence
of angiographic CAD. Similarly, case-control studies
in 51,051 patients, of whom 515 had CAD events,
found that 2 of the 5 common CRP haplotypes were
associated with significantly lower CRP levels, and
haplotype 4 was associated with significantly lower
CRP levels and an elevated risk for CAD. 32 The other
haplotypes were not associated with CAD. This find-
ing was not unexpected given that the identified ge-
netic polymorphisms predict a small proportion of
variation in hs-CRP, and most of the variation has not
been explained. 31 ,32 The apparent influence on base-
line CRP levels caused by CRP polymorphisms may
not be enough to alter CAD risk, which might explain
the lack of association between the majority of com-
mon haplotypes in the CRP gene and risk for CAD. 32
The PROVE IT-TIMI 22 tria1 9 found improved
outcomes with more aggressive treatment (atorvastat-
in 80 mg compared with pravastatin 40 mg).14 Signifi-
cantly lower event rates were observed in patients
with lower (<2 mg/L) versus higher (>2 mg/L) hs-CRP
concentrations at all achieved LDL-C concentrations
(2.8 vs 3.9 events per 100 person-years; P = 0.006).14
In fact, the lowest event rate (1.9 events per 100 person-
years; P < 0.001) was observed in patients achieving
both an LDL-C target <1.81 mmol/L «70 mg/dL) and
an hs-CRP target <2 mg/L. The A to Z trial,15 which
compared intensive and moderate treatment with sim-
vastatin (80 vs 20 mg), did not achieve the prespeci-
fied primary end point (composite of cardiovascular-
related death, nonfatal myocardial infarction,
readmission for ACS, and stroke); event rates were
14.4% in the 80-mg group and 16.7% in the simva-
statin 20-mg group at 6- to 24-month follow-up (HR,

December 2008 2307

 Clinical Therapeutics

Table II. Analysis of correlations between concentrations and changes from baseline (Ll) in hs-CRP and lipid
parameters after 26 weeks of treatment with atorvastatin 10 or 80 mg/d in patients with stable
coronary artery disease. (Spearman correlation coefficients were used.)

Atorvastatin 10 mg Atorvastatin 80 mg
(n = 170) (n = 169)

Concentration 6 Concentration 6
Lipid
Parameter r2 p r2 p r2 p r2 p

LDL-C 0.0324 0.016 0.0036 0.47 0.0484 0.005 0.0484 0.003

HDL-C 0.0169 0.080 0.0100 0.19 0.0004 0.770 0.0144 0.110
TG 0.0025 0.520 0.0081 0.24 0.0009 0.700 0.0004 0.820

LDL-C ~ low-density lipoprotein cholesterol; HDL-C ~ high-density lipoprotein cholesterol; TG ~ tnglycendes.

0.89; 95% CI, 0.76-1.04; P = NS). In a post hoc
analysis of results from the A to Z trial, hs-CRP con-
centrations that were strongly and independently as-
sociated with long-term survival were reached at
30 days and 4 monthsY After adjustment for other
risk factors, patients with hs-CRP >3 mg/L at 30 days
had significantly higher 2-year mortality rates than
those with hs-CRP 1 to 3 mg/L or hs-CRP <1 mg/L
(6.1 % vs 3.7% vs 1.6%; P < 0.001). Moreover, in
contrast with the significant early coronary event re-
duction found in PROVE IT-TIMI 22, the A to Z trial
did not find any significant between-group differences
in the primary end point during the first 4 months.
Despite a similar LDL-C concentration reached early

D Atorvastatln 10 mg (n - 170)
D Atorvastatln 80 mg(n -169)

100

80-

Vl
'-'
c 60-
.;:;
Q)

n:I
CL
4-
o 40-
Cf2.

20-
+
o-+-----'--------'---L.---,----I---'-------'--------'------,I
LDL-C <2.59 mmoljL LDL-C <1.81 mmoljL
«100 mgjdL) «70 mgjdL)

Figure 5. Proportions (95% CI) of patients achieving National Cholesterol Education Program Adult Treatment
Panel III low-density lipoprotein cholesterol (LDL-C) targets after 26-week treatment with atorvastatin
10 or 80 mg in patients with stable coronary artery disease (modified intent-to-treat population).

2308 Volume 30 Number 12


Table III. Tolerability with atorvastatin 10 and 80 mg/d
artery disease. Values are no. (%) of patients.
Atorvastatin
Parameter
All causality
~1 AE
~1 Musculoskeletal AEs
AST or ALT >3 x ULN
CK >10 x ULN
Treatment-related AEs
~1 AE
~1 Musculoskeletal AE
Discontinued due to AE
AE ~ adverse event; AST ~ aspartate aminotransferase; ALT
limit of normal; CK ~ creatinine kinase.
after treatment (1.61 mmol/L [62 mg/dL] at 120 days)
in the 2 trials, the hs-CRP reduction versus placebo
was >2-fold greater in PROVE IT-TIMI 22 13 com-
pared with that achieved in A to Z (38% vs 17%),15
which is possibly relevant to the difference in out-
comes between the 2 clinical trials.
The REVERSAL (Reversal of Atherosclerosis with
Aggressive Lipid Lowering) triaP3 found reduced ath-
erosclerosis progression rates (percentage change in
atheroma volume) after intensive (atorvastatin 80 mg/d,
-0.4%; 95% CI, -2.4% to 1.5%), compared with mod-
erate (pravastatin 40 mg/d, 2.7%; 95% CI,0.24-4.67;
P = NS), statin treatment in 502 patients with stable
CAD after 18 months using intravascular ultrasonog-
raphy. In REVERSAL, the reduction in atherosclerosis
progression was significantly related to greater reduc-
tions in the concentrations of lipoproteins and hs-
CRP (1" = 0.09; P = 0.04).34 The authors attributed this
additional benefit in part to the greater reduction in
hs-CRP in the atorvastatin group.
Taken together, the results from PROVE IT-TIMI
22, A to Z, and REVERSAL suggest that therapeutic
interventions to substantially lower CRP may have
ameliorative effects on coronary events and mortality.
We investigated the effects of low- versus high-dose
atorvastatin treatment on LDL-C in a population of
patients with stable CAD, a normal-range lipid profile,
December 2008
J. Bonnet et al.
In patients with stable coronary
Atorvastatin
10 mg 80 mg
(n = 170) (n = 170)
85 (50.0) 82 (48.2)
17(10.0) 12 (7.1)
0 3 (1.8)
1 (0.6) 1 (0.6)
14 (8.2) 20 (11.8)
6 (3.5) 6 (3.5)
3 (1.8) 6 (3.5)
~ alanine aminotransferase; ULN ~ upper
and evidence of low-grade systemic inflammation. The
population in the present study was similar to that in
the TNT (Treating to New Targets) study,35 which in-
cluded patients with clinically evident CAD and
slightly elevated LDL-C (<3.4 mmol/L [130 mg/dL])
and found a 22% relative risk reduction (HR, 0.78;
95% CI, 0.69-0.89; P < 0.001) with atorvastatin
80 mg/d compared with atorvastatin 10 mg/d.
This CAP study found that reductions in hs-CRP
were dose dependent. The maximum change in hs-CRP,
-25%, was achieved at week 5 in the 10-mg group.
In contrast, the change in hs-CRP at 5 weeks in the
80-mg group was -36.4%; this concentration de-
creased progressively over the 26-week study period,
to a total change of -57.1 %. These data are consistent
with the dose-dependent reductions, -27% in the
10-mg and -43% in the 80-mg group, in hs-CRP con-
centrations previously found with 12 weeks of treat-
ment with atorvastatin 10, 20, 40, and 80 mg/d in
628 patients with or without CAD in an RCT. 36 Simi-
lar results were reported in an RCT in 182 patients
with type 2 diabetes, in whom the median CRP con-
centrations were reduced by 15% and 47%, respec-
tively, with 30 weeks of treatment with atorvastatin
10 and 80 mg/d (P < 0.001).31 These findings suggest
that long-term treatment with a high-dose statin is nec-
essary to achieve maximum antiinflammatory effect.
2309
 Clinical Therapeutics
A post hoc analysis of data from the present study
found that the dual targets of LDL-C <1.81 mmoUL
«70 mg/dL) and hs-CRP <2.0 mg/L were reached
in a significantly greater proportion of patients with
atorvastatin 80 mg compared with less intensive treat-
ment with 10 mg (55.6% vs 13.5%; P < 0.001). How-
ever, the decrease in hs-CRP was largely independent
of baseline LDL-C and change in LDL-C and totally
independent of changes in HDL-C and TG, regardless
of dose. We report herein the correlation between the
magnitude of changes in hs-CRP and LDL-C, which
reached statistical significance with the 80-mg dose,
but not the 10-mg dose, of atorvastatin (80 mg, r2 =
0.0484; P = 0.003; 10 mg, r2 = 0.0036; P = NS). This
finding was similar to that from the PROVE IT-TIMI
22 trial 14 with atorvastatin 80 mg/d or pravastatin
40 mg/d (r = 0.16; P < 0.001). The greater correlation
between hs-CRP and LDL-C found in PROVE IT-
TIMI 22 compared with the present trial may be ac-
counted for by the comparatively smaller sample size,
large interindividual variability, different patient popu-
lation (eg, ACS), or perhaps site interaction in the
present trial. However, because <3 % of the variation
in achieved hs-CRP concentrations was explained by
the variation in achieved LDL-C in PROVE IT-TIMI
22, this small difference may not be clinically relevant.
Thus, statin-mediated reductions in hs-CRP may be
mediated by independent antiinflammatory effects of
inhibition of 3-hydroxy-3-methylglutaryl coenzyme A
reductase. 1O ,34,38--41 However, correlations of changes
in LDL-C and hs-CRP in individuals are decreased by
their measurement variability. A meta-analysis of
23 studies with 57 patients treated with a variety of
statins (atorvastatin, cerivastatin, lovastatin, pravasta-
tin, rosuvastatin, or simvastatin), nonstatin drugs, and
nondrug therapies found a strong correlation between
the changes in LDL-C and hs-CRP (r = 0.80;
P < 0.001), and concluded that statin therapies had no
significant effect on hs-CRP after adjustment for the
change in LDL-C. 42
The results from this CAP study suggest that the
80-mg dose of atorvastatin was generally well tolerat-
ed, with severe adverse events being reported in 2 pa-
tients, as a starting dose in patients with stable CAD
and low baseline LDL-C concentrations.
There were a number of limitations in the present
study. The population was restricted to patients with
at least moderately elevated hs-CRP concentrations
at baseline and did not include patients with severe
2310
hypercholesterolemia; thus, findings may not be ap-
plicable to the general population. A single hs-CRP
sample was taken at each time point, while the mean
of 2 samples taken 2 weeks apart would have been
preferable to account for variability in measure-
ments. 43 In addition, this study was not designed to
investigate the impact of lowering hs-CRP concentra-
tions on the incidence of coronary events.
It remains possible that underlying inflammatory pro-
cesses that predict coronary events cannot be captured
solely by variation in the CRP gene. The CAP study con-
tributes to the growing body of evidence supporting the
benefit of high-dose statin treatment in patients at high
risk, through LDL-C lowering and vascular benefits (hs-
CRP reduction), with the aim of reducing cardiovascular-
related mortality and morbidity in this population.
CONCLUSIONS
In this population of patients with stable CAD, normal-
range lipid profiles, and evidence of chronic low-grade
systemic inflammation, the effects of atorvastatin on
changes in hs-CRP were dose dependent, with the
high dose (80 mg) being associated with significantly
greater reductions in hs-CRP concentrations. Patients
in highest tertile of baseline hs-CRP experienced sig-
nificantly greater reductions in hs-CRP. Both low- and
high-dose atorvastatin were associated with a signifi-
cant and progressive decline in hs-CRP, largely inde-
pendent of changes in LDL-C, HDL-C, and TG. Ator-
vastatin 80 mg was generally well tolerated over the
study period.
ACKNOWLEDGM ENTS
This research and its publication were funded by
Pfizer France and Pfizer Canada.
We are grateful to the patients, local investigators,
and study coordinators and other study collabora-
tors who made completion of the study possible.
The study was designed and conducted by the CAP
steering committee members in collaboration with
Pfizer France and Pfizer Canada. The authors thank
Pauline Lavigne, MSc (CMED, Toronto, Ontario,
Canada), for her assistance with preparation of the
manuscript.
The authors had full access to the data and take
responsibility for its integrity. All authors have read
and agree to the manuscript as written.
The CAP Investigators included, from Canada:
P.T.S. Ma (Calgary, Alberta), A. Opgenorth (Edmon-
Volume 30 Number 12

ton, Alberta), J. Frohlich (Vancouver, British Colum-
bia), G. Hoag (Victoria, British Columbia), D. Mymin
(Winnipeg, Manitoba), E. Dr (Halifax, Nova Scotia),
D. Spence (London, Ontario), J.Y.M. Cha (Oshawa,
Ontario), R. McPherson (Ottawa, Ontario), M.J.
O'Mahony (Sarnia, Ontario), L.A. Leiter (Toronto,
Ontario), L.C.H. Yao (Weston, Ontario), D. Gaudet
(Chicoutimi, Quebec), R. Habib (Chomedey, Laval,
Quebec), c. Constance, J. Davignon, M.A. Lavoie, R.
Nadeau (Montreal, Quebec), c. Gagne (Sainte-Foy,
Quebec), P. Perron (Sherbrooke, Quebec), and T. Wil-
son (Saskatoon, Saskatchewan); from the Czech Re-
public: L. Berka (Jindrichuv Hradec), J. Bultas, M.
Herold, D. Zemanek (Prague), and V. Chaloupka (Br-
no-Bohunice); from France: P. Amarenco (Paris), P.
Assayag (Ie Kremlin Bicetre Cedex), M. Barboteu
(Evecquemont), J. Bonnet (Pessac Cedex), J. Boschat
(Brest); B. Citron (Clermont-ferrand), J.M. Davy
(Montpellier); E. Decoulx (Tourcoing), B. Doucet
(Chambery),J. Emmerich (Paris Cedex), D. Flammang
(Saint Michel), F. Funck (Pontoise), P. Guenoun (Thi-
onville), T. Haftel (Roubaix), J.c. Kahn (Poissy), K.
Khalife (Metz Cedex), P. Kohan (Nancy), J.M.
Lablanche (Lille), J. Lipiecki (Clermont-ferrand), M.
Martelet (Langres), T. Olive (Gap), J.P. Ollivier (Paris
Cedex), E. Perrier (Clamart), J. Puel (Toulouse), J.
Seoud (Creil), P. Voiriot (Vandoeuvre Les Nancy), and
J.E. Wolf (Dijon); from Poland: R. Bubinski (Bel-
chatow), R. Filipczak (Rawa Mazowiecka), K. Jawor-
ska (Torun),A. Kleinrok (Zamosc), K. Loboz-Grudzien
(Wroclaw), and M. Piepiorka (Gdynia); from Roma-
nia: M. Cinteza (Bucuresti), D. Dimulescu (Bucharest),
and M. Vintila (Bucuresti); from Russia: N. Ahmedzha-
nov (Moscow) and Y. Lopatin (Volgograd); from Slova-
kia: A. Dukat, S. Filipova, and J. Murin (Bratislava).
Dr. McPherson has received speaker and consulta-
tion fees from Pfizer Canada and has served on advi-
sory boards and received honoraria for Continuing
Medical Education from AstraZeneca Canada, Glaxo-
SmithKline, Merck Frosst Canada, Oryx Pharmaceu-
ticals, and Roche Pharmaceuticals. Dr. Davignon has
received research grant support from AstraZeneca,
Pfizer Canada, and Merck-Frosst Canada, and he is or
has been a consultant or scientific advisor for Astra-
Zeneca, Bayer, Bristol-Meyers Squibb, CardAlpha,
Cortia (Pharmena), Esperion, Genzyme Corporation,
Fournier/Solvay, Genfit, LifeCycle Pharma, McCain,
Merck, Merck/Frosst Schering, Merck-Schering, No-
vartis, Oryx, Pfizer Sankyo, sanofi-aventis, Takeda,
December 2008
J. Bonnet et al.
and Wyeth Pharmaceuticals. He is partlCIpating in
clinical trials for Pfizer, Merck, Pharmena, sanofi-
aventis, and Takeda and is a member of the speakers'
bureau of the International Atherosclerosis Society.
REFERENCES
1. Tedgu I A, Mallat Z. Cytokl nes In atherosclerosIs: Patho-
genic and regulatory pathways. Physiol Rev. 2006;86:515-
581.
2. Berk BC, Weintraub WS, Alexander RW. Elevation of
C-reactive protein In "active" coronary artery disease. Am
J Cardlol. 1990;65:168-172.
3. Lluzzo G, Biasucci LM, GalilmoreJR, et al. The prognostic
value of C-reactive protein and serum amyloid a protein In
severe unstable angina. N EnglJ Med. 1994;331 :417-424.
4. Verma S, DevaraJ S, Jlalal I. Is C-reactive protein an inno-
cent bystander or proatherogenlc culprit? C-reactive pro-
tein promotes atherothrombosls. Circulation. 2006;113:
2135-2150.
5. Wdlerson JT, Rldker PM. Inflammation as a cardiovascu-
lar nsk factor. Circulation. 2004;109(Suppl 1):112-111 O.
6. Venugopal SK, DevaraJ S,Jlalall. Effect of C-reactive pro-
tein on vascular cells: EVidence for a prolnflammatory,
proatherogenlc role. Curr Opln Nephrol Hypertens. 2005; 14:
33-37.
7. Jlalal I, DevaraJ S, Venugopal SK. C-reactive protein: Risk
marker or mediator In atherothrombosls? Hypertension.
2004;44:6-11.
8. Balgent C, Keech A, Kearney PM, et ai, for the Cholesterol
Treatment Tnallsts' (CTT) Collaborators. Efficacy and
safety of cholesterol-Iowenng treatment: Prospective
meta-analysIs of data from 90,056 participants In 14 ran-
domlsed tnals of statlns [published corrections appear In
Lancet. 2005;366:1358; Lancet. 2008;371 :2084]. Lancet.
2005;366:1267-1278.
9. Downs JR, Clearfield M, Wels S, et al. Pnmary prevention
of acute coronary events with lovastatln In men and
women with average cholesterol levels: Results of AF-
CAPS/TexCAPS. Air Force/Texas Coronary AtherosclerosIs
Prevention Study.JAMA. 1998;279:1615-1622.
10. Rldker PM, Rlfal N, Clearfield M, et ai, for the Air Force/
Texas Coronary AtherosclerosIs Prevention Study Investi-
gators. Measurement of C-reactive protein for the target-
Ing of statln therapy In the pnmary prevention of acute
coronary events. N EnglJ Med. 2001 ;344: 1959-1965.
11. Sacks FM, Pfeffer MA, Moye LA, et ai, for the Cholesterol
and Recurrent Events Tnal Investigators. The effect of
pravastatln on coronary events after myocardial infarc-
tion In patients with average cholesterol levels. N Engl
J Med. 1996;335:1001-1009.
12. Rldker PM, Rlfal N, Pfeffer MA, et ai, for the Cholesterol
and Recurrent Events (CARE) Investigators. Long-term ef-
2311
 Clinical Therapeutics
fects of pravastatln on plasma con-
centration of C-reactive protein. Cir-
culatlon.1999;100:230-235.
13. Cannon CP, Braunwald E, McCabe
CH, et ai, for the Pravastatln or
Atorvastatln Evaluation and Infec-
tion Therapy-ThrombolysIs In Myo-
cardlallnfarctlon 22 (PROVE IT-TIM I
22) Investigators. Intensive versus
moderate lipid lowenng with statlns
after acute coronary syndromes
[published correction appears In
N EnglJ Med. 2006;354:778]. N Engl
J Med. 2004;350: 1495-1504.
14. Rldker PM, Cannon CP, Morrow D,
et ai, for the Pravastatln or Ator-
vastatln Evaluation and Infection
Therapy-ThrombolysIs In Myocardial
Infarction 22 (PROVE IT-TIM I 22)
Investigators. C-reactive protein lev-
els and outcomes after statln thera-
py. N En~J Med. 2005;352:20-28.
15. de Lemos JA, Blazing MA, WIVIOtt
SD, et ai, for the A to Z Investiga-
tors. Early intensive vs a delayed
conservative slmvastatln strategy In
patients with acute coronary syn-
dromes: Phase Z of the A to Z tnal.
JAMA. 2004;292:1307-1316.
16. WIVIOtt SD, de Lemos JA, Cannon CP,
et al. A tale of two tnals: A com-
panson of the post-acute coronary
syndrome llpld-lowenng tnals A to
Z and PROVE IT-TIMI 22. Circula-
tlon.2006;113:1406-1414.
17. Morrow DA, de Lemos JA, Sabatine
MS, et al. Clinical relevance of C-
reactive protein dunng follow-up of
patients with acute coronary syn-
dromes In the Aggrastat-to-Zocor
Tnal. Circulation. 2006;114:281-288.
18. Schwartz GG, Olsson AG, Ezekow-
Itz MD, et ai, for the Myocardial
Ischemia Reduction with Aggressive
Cholesterol Lowenng (MIRACL)
Study Investigators. Effects of ator-
vastatln on early recurrent Ischemic
events In acute coronary syndromes:
The MIRACL study: A randomized
controlled tnal. JAMA. 2001 ;285:
1711-1718.
19. Klnlay S, Schwartz GG, Olsson AG,
et ai, for the Myocardial Ischemia
2312
Reduction with Aggressive Choles-
terol Lowenng Study Investigators.
High-dose atorvastatl n en hances
the decline In Inflammatory mark-
ers In patients with acute coronary
syndromes In the M IRACL study.
Circulation. 2003;108:1560-1566.
20. Meredith KG, Horne BD, Pearson
RR, et al. Companson of effects of
high (80 mg) versus low (20 mg)
dose of slmvastatln on C-reactive
protein and lipoproteins In patients
with anglographlc eVidence of coro-
naryartenal narrowing. AmJ Cardlol.
2007;99: 149-153.
21. SCinca BM, Morrow DA. Is C-reactive
protein an Innocent bystander or
proatherogenlc culpnt? The verdict
IS still out. Circulation. 2006; 113:
2128-2134; discussion 2151.
22. World Medical Association Decla-
ration of Helsinki: Ethical pnnclples
for medical research involving hu-
man subjects. JAMA. 2000;284:
3043-3045.
23. Sachs L. Applted StatistiCS: A Hand-
book of Techniques. 2nd ed. New
York, NY: Spnnger-Verlag; 1984.
24. Winer BJ, Brown DR, Michels KM.
Statistical PnnClples In Expenmental De-
sign. 3rd ed. New York, NY: McGraw-
Hill; 1991.
25. The Expert Panel on Detection, Evalu-
ation, and Treatment of High Blood
Cholesterol In Adults. Executive
summary of The Third Report of
the National Cholesterol Education
Program (NCEP) Expert Panel on
Detection, Evaluation, and Treat-
ment of High Blood Cholesterol In
Adults (Adult Treatment Panel III).
JAMA. 2001 ;285:2486-2497.
26. Grundy SM, Cleeman JI, Merz CN, et
ai, for the National Heart, Lung, and
Blood Institute; the Amencan Col-
lege of Cardiology Foundation; and
the Amencan Heart Association. Im-
plications of recent clinical tnals for
the National Cholesterol Education
Program Adult Treatment Panel III
gUidelines [published correction ap-
pears In Circulation. 2004;110:763].
Circulation. 2004;110:227-239.
27. LJbby P, Rldker PM. Inflammation and
atherosclerosIs: Role of C-reactive
protein In nsk assessment. AmJ Med.
2004;116(Suppl 6A):9S-16S.
28. Torzewskl J, Torzewskl M, Bowyer
DE, et al. C-reactive protein fre-
quently colocallzes with the termi-
nal complement complex In the in-
tima of early atherosclerotic lesions
of human coronary artenes. Arteno-
seier Thromb Vasc Bioi. 1998; 18:
1386-1392.
29. Yaron G, Bnll A, Dashevsky 0, et al.
C-reactive protein promotes platelet
adhesion to endothelial cells: A po-
tential pathway In atherothrombosls.
BrJ Haematol. 2006;134:426-431.
30. Song CJ, Nakagoml A, Chandar S,
et al. C-reactive protein contnbutes
to the hypercoagulable state In coro-
nary artery disease. J Thromb Hae-
most. 2006;4:98-106.
31. Grammer TB, Marz W, Renner
W, et al. C-reactive protein geno-
types associated with circulating C-
reactive protein but not with anglo-
graphic coronary artery disease:
The LURIC study [published online
ahead of pnnt May 21, 2008]. Eur
Heart}.
32. Pal JK, Mukamal KJ, Rexrode KM,
Rlmm EB. C-reactive protein (CRP)
gene polymorph Isms, CRP levels,
and nsk of incident coronary heart
disease In two nested case-control
studies. PLoS ONE. 2008;3:e1395.
33. Nissen SE, Tuzcu EM, Schoenhagen P,
et ai, for the REVERSAL Investiga-
tors. Effect of intensive compared
with moderate Iipld-lowenng therapy
on progression of coronary athero-
sclerosIs: A randomized controlled
tnal.JAMA. 2004;291:1071-1080.
34. Nissen SE, Tuzcu EM, Schoenhagen P,
et ai, for the Reversal of Atheroscle-
roSIS with Aggressive Lipid Lowenng
(REVERSAL) Investigators. Statl n
therapy, LDL cholesterol, C-reactive
protein, and coronary artery dis-
ease. N Engl J Med. 2005;352:29-
38.
35. LaRosaJC, Grundy SM, Waters DD,
et ai, forthe Treating to New Targets
Volume 30 Number 12
 J. Bonnet et al.

(TNT) Investigators. Intensive lipid ease Control and Prevention and A statement for health care profes-
lowenng with atorvastatln In pa- the Amencan Heart Association. sionals from the Centers for Disease
tients with stable coronary disease. Markers of inflammation and car- Control and Prevention and the
N Englj Med. 2005;352:1425-1435. diovascular disease: Application to Amencan Heart Association. Circu-
36. Ballantyne CM, Houn J, Notarbar- clinical and public health practice: latlon.2003;107:499-511.
tolo A, et ai, for the Ezetlmlbe Study
Group. Effect of ezetlmlbe co-
administered with atorvastatln In
628 patients with pnmary hyper-
cholesterolemia: A prospective, ran-
domized, double-blind trial. Circula-
tion. 2003; 107:2409-2415.
37. van de Ree MA, HUisman MV, Pnn-
cen HM, et ai, for the DALI-Study
Group. Strong decrease of high sen-
SitiVity C-reactive protein with hlgh-
dose atorvastatln In patients with
type 2 diabetes mellitus. Atheroscle-
roSIS. 2003; 166:129-135.
38. Rldker PM, Rlfal N, Lowenthal SP.
Rapid reduction In C-reactive pro-
tein with cenvastatln among 785 pa-
tients with pnmary hypercholester-
olemia. Circulation. 2001;103:1191-
1193.
39. Klnlay S, Timms T, Clark M, et ai,
for the Vascular BasIs Study Group.
Companson of effect of intensive
lipid lowenng with atorvastatln to
less Intensive lowenng with Iova-
statln on C-reactive protein In pa-
tients with stable angina pectons
and inducible myocardial Ischemia.
Am} Cardlol. 2002;89:1205-1207.
40. Jlalal I, Stein D, Balls D, et al. Effect
of hydroxymethyl glutaryl coenzyme
A reductase Inhlbltortherapy on high
sensitive C-reactive protein levels.
CirculatIOn. 2001;103:1933-1935.
41. PlengeJK, HernandezTL, Well KM,
et al. Slmvastatln lowers C-reactive
protein within 14 days: An effect
Independent of low-density lipo-
protein cholesterol reduction. Circu-
latlon.2002;106:1447-1452.
42. Klnlay S. Low-density Ilpoproteln-
dependent and -Independent e~
fects of cholesterol-Iowenng thera-
pies on C-reactive protein: A meta-
analysIs.} Am Call Cardlol. 2007;49: Address correspondence to: Jean Davignon, MD, Hyperlipidemia and
2003-2009. Atherosclerosis Research Group, Clinical Research Institute of Montreal
43. Pearson TA, Mensah GA, Alexander (IRCM), 110 Pine Avenue West, Montreal, Quebec, Canada H2W lR7.
RW, et ai, for the Centers for Dls- E-mail: davignj@ircm.qc.ca

December 2008 2313