Preventive Cardiology

Transcription

1 Preventive Cardiology C-Reactive Protein, but not Low-Density Lipoprotein Cholesterol Levels, Associate With Coronary Atheroma Regression and Cardiovascular Events After Maximally Intensive Statin Therapy Rishi Puri, MBBS; Steven E. Nissen, MD; Peter Libby, MD; Mingyuan Shao, MS; Christie M. Ballantyne, MD; Phillip J. Barter, MB, BS, PhD; M. John Chapman, PhD, DSc; Raimund Erbel, MD; Joel S. Raichlen, MD; Kiyoko Uno, MD, PhD; Yu Kataoka, MD; Stephen J. Nicholls, MBBS, PhD Background Baseline C-reactive protein (CRP) levels predict major adverse cardiovascular events (MACE: death, myocardial infarction, stroke, coronary revascularization, and hospitalization for unstable angina). The association between changes in CRP levels with plaque progression and MACE in the setting of maximally intensive statin therapy is unknown. Methods and Results The Study of Coronary Atheroma by Intravascular Ultrasound: Effect of Rosuvastatin Versus Atorvastatin (SATURN) used serial intravascular ultrasound measures of coronary atheroma volume in patients treated with rosuvastatin 40 mg or atorvastatin 80 mg for 24 months. The treatment groups did not differ significantly in the change from baseline of percent atheroma volume on intravascular ultrasound, CRP-modulating effects, or MACE rates, thus allowing for a (prespecified) post hoc analysis to test associations between the changes in CRP levels with coronary disease progression and MACE. Patients with nonincreasing CRP levels (n=621) had higher baseline (2.3 [ ] versus 1.1 [ ] mg/l; P<0.001) and lower follow-up CRP levels (0.8 [ ] versus 1.6 [ ] mg/l; P<0.001) versus those with increasing CRP levels (n=364). Multivariable analysis revealed a nonincreasing CRP level to independently associate with greater percent atheroma volume regression (P=0.01). Although the (log) change in CRP did not associate with MACE (hazard ratio, 1.18; 95% confidence interval, ; P=0.17), the (log) on-treatment CRP associated significantly with MACE (hazard ratio, 1.28; 95% confidence interval, ; P=0.02). On-treatment low-density lipoprotein cholesterol levels did not correlate with MACE (hazard ratio, 1.09; 95% confidence interval, ; P=0.45). Conclusions Following 24 months of potent statin therapy, on-treatment CRP levels associated with MACE. Inflammation may be an important driver of residual cardiovascular risk in patients with coronary artery disease despite aggressive statin therapy. Clinical Trial Registration URL: Unique identifier: NCT (Circulation. 2013;128: ) Key Words: atherosclerosis atorvastatin cholesterol, LDL C-reactive protein inflammation rosuvastatin ultrasonography Statin-mediated cholesterol lowering improves clinical outcomes. Despite this improvement, many individuals demonstrate atheroma progression, 1 and many such patients experience further cardiovascular events. Randomized clinical trials using the best contemporary medical therapies Clinical Perspective on p 2403 demonstrate, at most, relative risk reductions of the order of 30% to 40%. This highlights a significant degree of residual cardiovascular risk experienced by a large proportion of individuals prescribed statins for secondary clinical prevention. 2 Accordingly, there is a need to identify novel targets and therapeutic strategies to effectively combat this residual risk. Received June 3, 2013; accepted August 29, 2013 From the Department of Cardiovascular Medicine, Cleveland Clinic, Cleveland, OH (R.P., S.E.N., K.U., Y.K., S.J.N.); Cardiovascular Division, Brigham and Women s Hospital, Boston, MA (P.L.); C5Research, Cleveland Clinic, Cleveland, OH (M.S.); Section of Cardiovascular Research, Baylor College of Medicine, and the Methodist DeBakey Heart and Vascular Center, Houston, TX (C.M.B.); Centre for Vascular Research, University of New South Wales, Sydney, Australia (P.J.B.); INSERM Dyslipidaemia and Atherosclerosis Research Unit, Hȏpital de la Pitié, Paris, France (M.J.C.); West German Heart Center, Essen, Germany (R.E.); AstraZeneca, Wilmington, DE (J.S.R.); and South Australian Health and Medical Research Institute, University of Adelaide, Adelaide, Australia (S.J.N.). Guest editor for this article was Daniel J. Rader, MD. Correspondence to Stephen J. Nicholls, MBBS, PhD, South Australian Health and Medical Research Institute, Level 9, 121 King William St, Adelaide, SA, 5001, Australia. stephen.nicholls@sahmri.com 2013 American Heart Association, Inc. Circulation is available at DOI: /CIRCULATIONAHA

2 2396 Circulation November 26, 2013 Multiple avenues of research implicate inflammation as a factor involved in the initiation, progression, and instability of atherosclerotic plaques. 3,4 Elevations of serum inflammatory biomarkers consistently associate with the risk of experiencing a cardiovascular event. 5,6 C-reactive protein (CRP), in particular, has been extensively investigated, and the prognostic utility of baseline CRP levels is now considered by some to be equivalent to that of serum cholesterol levels. 7,8 Although many benefits of statins are attributed to cholesterol lowering, both coronary atheroma regression 9 and clinical event reductions 10 have also been shown independently to correlate with statin-mediated CRP lowering. Yet, there remains an ongoing debate as to whether the effects of statin-mediated CRP lowering are simply a function of parallel changes in low-density lipoprotein cholesterol (LDL-C) or follow a separate mechanistic pathway. The prognostic benefit of CRP in statin-treated patients who achieve LDL-C levels <70 mg/dl is uncertain. 11 The Study of Coronary Atheroma by Intravascular Ultrasound: Effect of Rosuvastatin Versus Atorvastatin (SATURN; clinicaltrials.gov unique identifier NCT ) was the first head-to-head comparison of the antiatherosclerotic effect of 2 potent statin agents, each prescribed at their maximal approved dose, on coronary atherosclerosis progression. 12 Treatment with both agents was associated with the regression of coronary atheroma, although there was no significant difference between the 2 groups in the primary efficacy end point of the change in percent atheroma volume (PAV); nor was there a difference in safety or clinical event rates between the treatment groups. The primary objective of this prespecified post hoc analysis was to examine the prognostic effect of changes in CRP levels on both coronary atheroma progression and major adverse cardiovascular events (MACE) in patients following 24 months of maximally intensive statin therapy. Methods Patient Selection The design of SATURN has been previously described. 13 In brief, patients with angiographically demonstrable coronary disease and LDL-C <116 mg/dl following a 2-week treatment period with atorvastatin 40 mg or rosuvastatin 20 mg daily were re-randomly assigned and treated for 24 months with atorvastatin 80 mg or rosuvastatin 40 mg daily. Subjects underwent intravascular ultrasound (IVUS) imaging of a coronary artery at baseline and after 104 weeks of treatment. Acquisition and Analysis of Intravascular Coronary Imaging The presence of at least a single lumen stenosis of >20% angiographic diameter stenosis severity in an epicardial coronary artery at the time of a clinically indicated coronary angiogram was necessary for enrolment eligibility. IVUS was performed at baseline in a single, native coronary artery with no lumen stenosis of >50% severity that had not undergone revascularization and was not considered to be the culprit vessel of a previous myocardial infarction. Images were screened by the Atherosclerosis Imaging Core Laboratory of the Cleveland Clinic Center for Clinical Research for quality, and those patients whose baseline imaging met these requirements were eligible for randomization. Following 104 weeks of treatment, patients underwent a second IVUS of the same artery. Anatomically matched arterial segments were selected for analysis on the basis of proximal and distal landmarks. Cross-sectional images spaced 1 mm apart were selected for analysis, with lumen and external elastic membrane leading edges defined by manual planimetry. Plaque area was determined as the area between these leading edges. PAV, a measure of plaque burden, was calculated as previously described. 14 Change in plaque burden was calculated as the PAV at 104 weeks minus the corresponding PAV at baseline. Plaque regression was defined as any decrease in PAV from baseline. The post hoc analyses presented here pooled results from both treatment groups; as in SATURN, they did not differ in the primary efficacy end point of the change in PAV from baseline, and the average on-treatment CRP levels were not different between the rosuvastatin- and atorvastatin-treated groups. Statistical Analysis Continuous variables were reported as mean±standard deviation if normally distributed and as median (interquartile range) if nonnormally distributed. Demographics, baseline clinical characteristics, follow-up medications, and baseline ultrasonic and laboratory biochemical data were compared between those who had an increase in average on-treatment CRP levels from baseline, against those whose on-treatment CRP levels did not increase. Two-sample t tests were used for normally distributed continuous variables, Wilcoxon rank sum tests were used for nonnormally distributed continuous variables, and χ 2 tests were used for categorical variables. Serial changes in IVUS measurements were analyzed by analysis of variance, adjusting for their baseline counterparts, and were reported as least-squares mean±standard error. Given the significant difference observed for change in PAV between those whose CRP increased versus those whose CRP did not increase, a multivariable analysis of variance analysis was undertaken to assess for its association with changes in CRP levels after controlling for clinical factors. Partial correlation analyses were performed to assess for relationships between change in CRP and lipid parameters, or change in PAV. A multivariable logistic regression model was constructed for defining the important predictors of any reduction in CRP from baseline. A multivariable Cox proportional hazards model was derived to identify clinical factors that were associated with MACE (defined as cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, coronary revascularization, or hospitalization for unstable angina), along with examining the effect of CRP (change in CRP and on-treatment CRP levels) on MACE. To create these multivariable models, demographic data and clinical characteristics were entered into a multivariable logistic or Cox model (with CRP forced into this model) for variable selection with bootstrap resampling (1000 iterations and a P value criterion of 0.1 for retention). Those variables having a 40% probability of retention were entered into a second logistic or Cox model with the stepwise model selection procedure. The significance level to enter and keep a variable was set at For the Cox model, the risk factors met the proportional hazards assumption by plotting the log of the negative log of their estimated survival distribution, log( log(s(t))), versus time, and the linearity assumption was assessed and satisfied for all continuous variables by plotting the log hazard ratio against the continuous variable, and testing the linear hypotheses, as well. The selected covariates formed the covariate set for the final multivariable logistic model for on-treatment CRP or the final multivariable Cox model for MACE. Given the significant association between on-treatment CRP levels and MACE, further evaluations were conducted by performing log-rank tests with Kaplan Meier figures and unadjusted and adjusted Cox proportional hazards regressions with hazard ratio (HR) and associated 95% confidence interval with forest plots. To better understand the relationships between MACE, CRP, and LDL-C, similar analyses were conducted for changes in CRP and for on-treatment LDL-C. These survival analyses applied a cutoff date of 24 months and used the time to the first occurrence of MACE for analysis. Patients with no MACE by 24 months were censored at this time point. The MACE analysis included all patients who were randomly assigned according to the treatment regimen with on-treatment CRP records (n=1263). All remaining analyses used the modified intent-to-treat

3 Puri et al Statins, CRP, Plaque Progression, and MACE 2397 population, which included all patients with a baseline and follow-up IVUS measurement, and on-treatment CRP levels (n=985), as well. A 2-sided P value of <0.05 was considered statistically significant. All analyses were performed with the use of SAS software version 9.2 (SAS Institute, NC). Results Patient Characteristics Table 1 shows the baseline demographics, clinical characteristics, and concomitant medications of patients who had evaluable CRP levels, stratified according to whether CRP levels increased (n=364) or did not increase (n=621) from baseline. The mean age of the studied population was 57.6±8.5 years, with a male preponderance (73.4%). The 2 patient groups did not differ at baseline in their cardiovascular risk factor profile, mode of clinical presentation, or concomitant medications. Laboratory Measurements Table 2 shows baseline, follow-up, and changes of lipids and CRP levels. During the course of the study, LDL-C decreased by 44% from 120.3±28 mg/dl to 65.9±23 mg/dl, highdensity lipoprotein cholesterol (HDL-C) increased by 12% from 44.9±11mg/dL to 49.4±12 mg/dl, and CRP decreased by 33% from 1.6 ( ) mg/l to 1.1 ( ) mg/l. At baseline, patients with increasing CRP levels had higher HDL-C (46.1±12 mg/dl versus 44.2±11 mg/dl, P=0.02), and apolipoprotein A-1 (130.1±25 mg/dl versus 125.2±23 mg/dl, P=0.003) levels, and lower apolipoprotein B: apolipoprotein A-1 (0.83±0.23 versus 0.87±0.24, P=0.008) and CRP (1.1 mg/l [ ] versus 2.3 [ ], P<0.001) levels in comparison with those whose CRP levels did not increase. At follow-up, patients with increasing CRP levels demonstrated higher levels of LDL-C (67.5±22 mg/dl versus 64.9±22 mg/dl, P=0.04) and CRP (1.6 mg/l [ ] versus Table 1. Variable Demographics, Baseline Characteristics, and Concomitant Medications Total 0.8 [ ], p<0.001) in comparison with those whose CRP levels did not increase. Atheroma Burden and Vascular Dimensions Table 3 describes the baseline and change in IVUS measurements in patients stratified according to changes in CRP levels. In the overall population, PAV decreased by 1.23±0.09% (P<0.001), with 67% of patients showing regression, defined as any reduction in PAV from baseline. No difference in baseline PAV was observed between patients whose CRP levels did or did not increase (36.6±8.0% versus 36.0±8.4%, P=0.29). However, patients with increasing CRP levels demonstrated less PAV regression than those whose CRP levels did not increase ( 0.91±0.15% versus 1.42±0.11%, P=0.007). Lumen volume expanded in patients whose CRP levels did not increase (+2.30±1.17 mm 3, P=0.05 from baseline), but not in those with increasing CRP ( 1.00±1.53 mm 3, P=0.52 from baseline). External elastic membrane volumes decreased similarly in both groups. Multivariable analysis demonstrated that a lack of an on-treatment rise in CRP independently associated with PAV regression (P=0.01). Association of Changes in CRP Levels With Changes in Lipid Parameters After controlling for baseline CRP and lipid values, the change in CRP correlated weakly but significantly with changes in LDL-C (r=0.07, P=0.02), non-hdl-c (r=0.1, P=0.001), triglycerides (r=0.1, P=0.001), and apolipoprotein B levels (r=0.09, P=0.008). Multivariable Analysis of Factors Associated With Any Reduction in CRP From Baseline Table 4 summarizes a multivariable logistic regression analysis of factors associated with any reductions in CRP levels from baseline. A higher CRP level at baseline independently Change From Baseline CRP (n=985) No Increase (n=621) Increase (n=364) Age, y 57.6± ± ± Men 723 (73.4) 452 (72.8) 271 (74.5) 0.57 Body mass index, kg/m ± ± ± Previous MI 241 (24.5) 150 (24.2) 91 (25.0) 0.77 Hypertension 694 (70.5) 435 (70.0) 259 (71.2) 0.71 Diabetes mellitus 151 (15.3) 99 (15.9) 52 (14.3) 0.49 Current smoker 318 (32.3) 204 (32.9) 114 (31.3) 0.62 ACS at presentation 335 (34) 220 (35.4) 115 (31.6) 0.22 Rosuvastatin 493 (49.9) 301 (48.5) 192 (52.7) 0.20 Aspirin 607 (61.6) 384 (61.8) 223 (61.3) 0.86 β-blockers 602 (61.1) 382 (61.5) 220 (60.4) 0.74 ACE inhibitors 432 (43.9) 268 (43.2) 164 (45.1) 0.56 Angiotensin receptor blocker 163 (16.5) 98 (15.8) 65 (17.9) 0.40 Nitrates 821 (83.4) 520 (83.7) 301 (82.7) 0.67 Values of continuous variables are reported as mean±sd if normally distributed, and median (interquartile range) if not normally distributed. Categorical variables are reported as n (%). All listed medications are concomitant. ACS indicates acute coronary syndrome; ACE, angiotensin-converting enzyme; CRP, C-reactive protein; MI, myocardial infarction; and SD, standard deviation. P Value

4 2398 Circulation November 26, 2013 Table 2. Laboratory Findings Variable Total (n=985) No Increase (n=621) Change From Baseline CRP Increase (n=364) Baseline LDL-C 120.3± ± ± HDL-C 44.9± ± ± Non-HDL-C 149.0± ± ± Triglycerides 128 (94,180) 128 (93, 181) 129 (96, 177) 0.79 ApoB 105.3± ± ± ApoA ± ± ± ApoB:ApoA ± ± ± CRP 1.6 ( ) 2.3 ( ) 1.1 ( ) <0.001 Follow-up* LDL-C 65.9± ± ± HDL-C 49.4± ± ± Non HDL-C 91.2± ± ± Triglycerides 127.1± ± ± ApoB 73.9± ± ± ApoA ± ± ± ApoB:ApoA ± ± ± CRP 1.1 ( ) 0.8 ( ) 1.6 ( ) <0.001 Change from baseline LDL-C Mean 54.4± ± ± % change 43.5± ± ± HDL-C Mean 4.5± ± ± % change 11.6± ± ± Non-HDL-C Mean 57.9± ± ± % change 37.2± ± ± Triglycerides Mean 17.0± ± ± % change 0.9± ± ± ApoB Mean 31.5± ± ± % change 28.5± ± ± ApoA-1 Mean 14.8± ± ± % change 13.5± ± ± ApoB:ApoA-1 Mean 0.32± ± ±0.18 <0.001 % change 36.1± ± ±16.8 <0.001 CRP Median (IQR) 0.4 ( 1.6 to 0.2) 1.0 ( 2.7 to 0.4) 0.4 (0.0 to 1.9) <0.001 % change <0.001 Values of continuous variables are reported as mean±sd if normally distributed, and median (interquartile range) if not normally distributed. All lipid and lipoprotein measurements are in mg/dl, CRP measurements are mg/l. Apo indicates apolipoprotein; CRP, C-reactive protein; HDL-C, high-density lipoprotein cholesterol; IQR, interquartile range; LDL-C, low-density lipoprotein; and SD, standard deviation. *Unless otherwise noted, laboratory values obtained during treatment are the time-weighted averages of all postbaseline values. P Value

5 Puri et al Statins, CRP, Plaque Progression, and MACE 2399 Table 3. Ultrasonic Measures of Atheroma Volume and Vascular Dimensions IVUS Parameter Total Change From Baseline CRP (n=985) No Increase (n=621) Increase (n=364) Percent atheroma volume, % Baseline 36.4± ± ± Change from baseline 1.23± ± ± P value for change from baseline <0.001 <0.001 <0.001 Lumen volume, mm 3 Baseline 248.0± ± ± Change from baseline 1.08± ± ± P value for change from baseline EEM volume, mm 3 Baseline 392.2± ± ± Change from baseline 6.11± ± ± P value for change from baseline <0.001 <0.001 <0.001 Baseline and follow-up values are reported as mean±sd, and change values are reported as least squares mean±se. CRP indicates C-reactive protein; EEM, external elastic membrane; IVUS, intravascular ultrasound; SD, standard deviation; and SE, standard error. *P value reflects comparisons between groups dichotomized according to change in CRP level. P Value* correlated with a decrease of CRP levels from baseline. Higher baseline levels of triglycerides, HDL-C, LDL-C:HDL-C ratio, and body mass index, along with a greater change in the LDL- C:HDL-C ratio, were all associated with a greater likelihood of an increase in CRP level from baseline. CRP Levels and MACE Table 5 describes a multivariable Cox regression analysis of factors associated with MACE. After controlling for the use of angiotensin-converting enzyme inhibitors, diabetes mellitus, and baseline CRP, the change in CRP did not associate significantly with MACE (HR, 1.20; ; P=0.14). However, when controlling for these factors, a baseline clinical presentation with acute coronary syndrome, the on-treatment CRP level significantly and independently associated with MACE (HR, 1.29; ; P=0.01). In SATURN, 101 patients experienced 143 adjudicated MACE (4 cardiovascular deaths, 22 nonfatal myocardial infarctions, 5 nonfatal ischemic strokes, 29 hospitalizations for unstable angina, and 83 coronary revascularization procedures). For the purposes of this analysis, we additionally Table 4. Multivariable Logistic Regression Analysis for Any Reduction in CRP From Baseline Parameter Odds Ratio (95% CI) per SD P Value (Log) baseline CRP 2.78 ( ) < (Log) baseline triglyceride level 0.84 ( ) 0.03 Baseline BMI 0.83 ( ) 0.02 Baseline HDL-C level 0.82 ( ) 0.04 Baseline LDL-C:HDL-C ratio 0.68 ( ) 0.01 Change in LDL-C:HDL-C ratio 0.66 ( ) The multivariable logistic regression analysis controls for treatment and region. Odds ratios are reported per standard deviation of each parameter. BMI indicates body-mass index; CI, confidence interval; CRP, C-reactive protein; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein; and SD, standard deviation. defined hard MACE as including only the hard clinical end points of cardiovascular death, nonfatal myocardial infarction, and nonfatal stroke as a sensitivity analysis. Figure 1 shows a Cox proportional model for HRs of time to first MACE according to changes in CRP levels, on-treatment level of CRP, and on-treatment level of LDL-C. Following multivariable adjustment, the changes in CRP did not associate significantly with MACE (multivariable adjusted HR, 1.18; ; P=0.17); however, there was a significant association between on-treatment CRP levels and MACE (multivariable adjusted HR, 1.28; ; P=0.02). In contrast, on-treatment LDL-C levels did not significantly associate with MACE (multivariable adjusted HR, 1.09; ; P=0.45). For hard MACE, the multivariable adjusted HR for changes in CRP approached statistical significance (HR, 1.44; ; P=0.08); however, the association between on-treatment CRP levels and hard MACE did achieve marginal statistical significance (multivariable adjusted HR, 1.42; ; P=0.05). Table 5. Parameter Multivariable Cox Regression Analysis for MACE Hazard Ratio for MACE (95% CI) P Value Change in CRP Concomitant ACE inhibitor use 2.25 ( ) History of diabetes mellitus 1.79 ( ) 0.02 (Log) baseline CRP 1.28 ( ) 0.05 (Log) change in CRP 1.18 ( ) 0.17 On-treatment CRP Concomitant ACE inhibitor use 2.07 ( ) History of diabetes mellitus 1.97 ( ) History of acute coronary syndrome 1.60 ( ) 0.03 (Log) on-treatment CRP 1.28 ( ) 0.02 ACE indicates angiotensin-converting enzyme; CI, confidence interval; CRP, C-reactive protein; and MACE, major adverse cardiovascular event. Hazard ratios for log-transformed change in CRP, log-transformed baseline CRP and log-transformed on-treatment CRP are reported per standard deviation.

6 2400 Circulation November 26, 2013 Figure 1. Cox proportional hazard ratios for time to first MACE or hard MACE, according to the change from baseline of CRP levels, on-treatment CRP levels, or on-treatment LDL-C levels. *Values of log-transformed change in CRP, on-treatment CRP, and on-treatment LDL-C were per standard deviation. Adjusted for history of diabetes mellitus, concomitant ACE inhibitor use, and log-transformed baseline CRP. Adjusted for history of diabetes mellitus, history of acute coronary syndrome, and concomitant use of ACE inhibitor. Adjusted for history of diabetes mellitus, history of acute coronary syndrome, concomitant ACE inhibitor use, and log-transformed on-treatment CRP. ACE indicates angiotensin-converting enzyme; CI, confidence interval; CRP, C-reactive protein; HR, hazard ratio; LDL-C, low-density lipoprotein cholesterol; and MACE, major adverse cardiovascular events. Again, on-treatment LDL-C levels failed to associate significantly with hard MACE (multivariable adjusted HR, 1.20; ; P=0.35). Figure 2 shows a Kaplan Meier survival curve analysis according to quartiles of on-treatment CRP levels and MACE. This shows a stepwise relationship between increasing on-treatment CRP levels and MACE. For on-treatment CRP levels >2.8 mg/l (quartile 4), the 2-year cumulative incidence of MACE was 9.5% (log-rank P=0.07 against all lower quartiles of on-treatment CRP). MACE rates for quartiles 1 (CRP mg/l), 2 ( mg/l), and 3 ( mg/l) were 5.9%, 6.6%, and 7.6%, respectively. Discussion The current analysis explored the effect of changes in CRP levels on coronary atheroma progression and cardiovascular events in patients who achieved very low LDL-C levels (on average, 65.9 mg/dl) following 2 years of maximally intensive statin therapy. Despite potent statin therapy, 37% of individuals experienced rising on-treatment CRP levels from baseline, and changes in LDL-C and CRP correlated weakly. Decreasing CRP levels independently associated with coronary atheroma regression. Although the absolute change in CRP was nonprognostic, the on-treatment level of CRP independently associated with MACE. Furthermore, a stepwise relationship was noted between on-treatment CRP levels and time to first MACE. Collectively, these results highlight the potential importance of on-treatment CRP levels mediating residual cardiovascular risk in individuals with coronary artery disease treated according to current lipid-lowering guidelines. For decades, circulating cholesterol levels have justifiably received major attention as a modifiable biomarker of cardiovascular risk. Statin treatment meaningfully reduces cholesterol levels in humans, and lowers cardiovascular events in individuals in broad categories of risk. 15 Individuals treated with statins who achieve lowering of both LDL-C and CRP experience the best clinical outcomes, 10,16 18 indicating the clinical importance of the Figure 2. Kaplan Meier survival curve analysis of quartiles of on-treatment CRP levels and MACE. The cumulative incidence of MACE in patients with the lowest quartile of on-treatment CRP was 5.9% in comparison with 9.5% in those with the highest quartile of baseline coronary PAV (P=0.06). CRP indicates C-reactive protein; MACE, major adverse cardiovascular events; and PAV, percent atheroma volume.

7 Puri et al Statins, CRP, Plaque Progression, and MACE 2401 anti-inflammatory and LDL-lowering effects, as well, of this drug class. A recent meta-analysis of 52 prospective studies, inclusive of nearly patients, confirmed numerous previous prospective observations of the independent prognostic capacity of CRP for first-ever cardiovascular events, of the same order as LDL-C, HDL-C, and systolic blood pressure. 7 A 1 standard deviation increase in CRP was associated with a multivariable adjusted HR for a cardiovascular event of 1.20, a value that compared favorably with the HR for a cardiovascular event of 1.17 per 1 standard deviation increase in total cholesterol. Our current findings extend these previous observations by examining the relationship between changes and on-treatment CRP levels, and serial rates of coronary atheroma progression on IVUS, and cardiovascular events in patients treated with maximally intensive statin therapy, as well. Patients enrolled in SATURN achieved one of the lowest on-treatment LDL-C levels of any previous atherosclerosis imaging trial. Despite this finding, each 1 standard deviation increase in CRP associated with a multivariable adjusted HR for MACE of 1.28 (P=0.02) and for hard MACE of 1.42 (P=0.05). This finding suggests that, in patients with atherosclerotic disease, systemic inflammation could represent an important prognostic biomarker of residual cardiovascular risk. This will ultimately require prospective confirmation in a large-scale clinical outcomes trial. This study found a significant difference in the magnitude of change in coronary atheroma volume according to change in CRP levels. In the Reversal of Atherosclerosis with Aggressive Lipid Lowering (REVERSAL) trial, the intensively treated patient group demonstrated reduced coronary atheroma progression in comparison with patients who received less intense treatment, 19 especially in patients who achieved reductions of both LDL-C and CRP below the median. 9 The observation in SATURN of greater coronary atheroma regression and lumen expansion in patients whose CRP levels did not increase, in comparison with those whose CRP levels increased, provides further evidence in support of the role of inflammation in plaque evolution. Many consider statin-mediated CRP lowering, and the corresponding anti-inflammatory effects, to result from the change in LDL-C levels. 20 As such, the relevance of CRP to clinical outcomes in statin-treated patients has engendered some controversy. In REVERSAL, correlation analysis revealed that reductions in atherogenic lipoprotein levels correlated weakly with CRP reductions (change in LDL-C versus change in CRP: r=0.13, P=0.005), with no correlations between changes in LDL-C and CRP noted in each treatment group alone. 9 Changes in LDL-C correlated negligibly with an individual s change in CRP in numerous large outcome trials including the Cholesterol and Recurrent Events (CARE) trial, 21 PROVE-IT, 22 and Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin (JUPITER). 17 In the overall SATURN population, we similarly report only a weak correlation between changes in CRP and changes in LDL-C (r=0.07, P=0.02). Previous observations also show statin-mediated CRP lowering noted to occur rapidly and independently of LDL-C lowering. 23 The consistent independent variation of reductions in LDL-C and CRP in individuals treated with statins lends further support to the notion that statin-mediated CRP reductions do not depend merely on LDL-C lowering. In the present analysis, on-treatment CRP levels associated with MACE, including a trend for an association with hard MACE. It is important to note that our findings do not lend support for a causal role of CRP in the genesis of cardiovascular events. 24 Rather, our findings add to the considerable evidence supporting the prognostic role of persistent inflammation, 25 even in individuals with coronary artery disease treated optimally with potent statin regimens and very low achieved LDL-C levels. These findings also have important implications for the quest to develop novel therapeutic strategies specifically targeting inflammation, to possibly achieve added risk reduction in those patients already optimized with maximally intensive statin therapy. Furthermore, the observed trend toward a stepwise relationship between on-treatment CRP levels and risk of MACE suggests that therapies designed to lower inflammation could be titrated. 26 This implication may ultimately depend on the magnitude of CRP lowering required to achieve net clinical benefit. This analysis has limitations. Our findings are a prespecified post hoc analysis of a clinical trial that failed to meet its primary efficacy end point of elucidating differences in coronary atheroma volume progression between the 2 potent statin regimens. However, the treatment assignments in SATURN represent the most intensive statin regimens currently available, and SATURN demonstrated no differences in the magnitude of PAV regression (primary efficacy end point), CRP-lowering capabilities, or clinical outcome between these 2 agents. Nevertheless, these results should be considered as hypothesis generating. Furthermore, other unmeasured confounding variables could have influenced systemic levels of inflammation, biasing our results. We also cannot attribute plaque and vessel responses measured on IVUS specifically to the MACE rates observed in individuals with higher on-treatment CRP levels. However, similar to the deleterious effects of systemic inflammation on the arterial wall, atherosclerosis is a systemic disease, and common to most atherosclerosis-imaging trials, the imaged vessel simply represents a snapshot of the serial response of the arterial vasculature to such conditions. In conclusion, in patients with coronary artery disease, most of whom achieved LDL-C levels <70 mg/dl (average, 65.9 mg/dl) following 2 years of maximally intensive statin therapy; over one-third of patients did not experience a fall in on-treatment CRP from baseline. These patients demonstrated less coronary atheroma regression than those with decreasing CRP levels. On-treatment CRP, but not on-treatment LDL-C levels, independently associated with MACE and displayed a strong trend toward associating with hard MACE. These findings suggest that systemic inflammation may contribute to cardiovascular risk in optimally treated individuals, and they underscore the importance of clinical trials that will specifically test the inflammatory hypothesis of atherothrombosis.

8 2402 Circulation November 26, 2013 Acknowledgments We acknowledge the technical expertise of the Atherosclerosis Imaging Core Laboratory, Cleveland Clinic Coordinating Center for Clinical Research (C5R). Sources of Funding SATURN was funded by AstraZeneca. Disclosures Dr Nissen has received research support to perform clinical trials through the Cleveland Clinic Coordinating Center for Clinical Research from Pfizer, AstraZeneca, Novartis, Roche, Daiichi- Sankyo, Takeda, Sanofi-Aventis, Resverlogix, and Eli Lilly; and is a consultant/advisor for many pharmaceutical companies but requires them to donate all honoraria or consulting fees directly to charity so that he receives neither income nor a tax deduction. Dr Libby serves as an unpaid consultant for Novartis, Johnson & Johnson, Amgen, and Roche; serves in unpaid leadership roles for clinical trials sponsored by AstraZeneca, GlaxoSmithKline, Novartis, Pfizer, Pronova, and Sigma Tau; and previously received royalties from Roche for the patent on CD40L in cardiovascular risk stratification. Dr Ballantyne has received grant support from Abbott, Astra-Zeneca, Bristol-Myers Squibb, Genentech, GlaxoSmithKline, Kowa, Merck, Novartis, Roche, Sanofi-Synthelabo, and Takeda; consulting fees and honoraria from Abbott, Adnexus, Amarin, Amylin, AstraZeneca, Bristol-Myers Squibb, Esperion, Genentech, GlaxoSmithKline, Idera, Kowa, Merck, Novartis, Omthera, Resverlogix, Roche, Sanofi-Synthelabo, and Takeda; lecture fees from Abbott, AstraZeneca, GlaxoSmithKline, and Merck. Dr Barter holds an advisory board position for AstraZeneca, Merck, Roche, CSL Behring, and Pfizer; receives grant support from Merck; consulting fees from CSL Behring; lecture fees from AstraZeneca, Kowa, Merck, Pfizer, and Roche. Dr Chapman receives grant support from CSL, GlaxoSmithKline, Pfizer, Randox, and Merck; consulting fees from Merck and Pfizer; lectures fees from Genzyme, Merck, Lilly, Roche, and Kowa; equity interests in Wiley Inc. Dr Erbel receives grant and travel support from the Heinz Nixdorf Foundation, German Research Foundation; accommodations/meeting expenses from Biotronik, Sanofi, and Novartis. Dr Raichlen is an employee of and owns stock in AstraZeneca. Dr Nicholls has received speaking honoraria from AstraZeneca, Pfizer, Merck Schering- Plough, and Takeda; consulting fees from AstraZeneca, Pfizer,Merck Schering-Plough, Takeda, Roche, NovoNordisk, LipoScience, and Anthera; research support from AstraZeneca and Lipid Sciences. The other authors report no conflicts. References 1. Bayturan O, Kapadia S, Nicholls SJ, Tuzcu EM, Shao M, Uno K, Shreevatsa A, Lavoie AJ, Wolski K, Schoenhagen P, Nissen SE. Clinical predictors of plaque progression despite very low levels of low-density lipoprotein cholesterol. J Am Coll Cardiol. 2010;55: Libby P. The forgotten majority: unfinished business in cardiovascular risk reduction. J Am Coll Cardiol. 2005;46: Ross R. Atherosclerosis an inflammatory disease. N Engl J Med. 1999;340: Libby P. Inflammation in atherosclerosis. Arterioscler Thromb Vasc Biol. 2012;32: Liuzzo G, Biasucci LM, Gallimore JR, Grillo RL, Rebuzzi AG, Pepys MB, Maseri A. The prognostic value of C-reactive protein and serum amyloid a protein in severe unstable angina. N Engl J Med. 1994;331: Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH. Inflammation, aspirin, and the risk of cardiovascular disease in apparently healthy men. N Engl J Med. 1997;336: Kaptoge S, Di Angelantonio E, Pennells L, Wood AM, White IR, Gao P, Walker M, Thompson A, Sarwar N, Caslake M, Butterworth AS, Amouyel P, Assmann G, Bakker SJ, Barr EL, Barrett-Connor E, Benjamin EJ, Bjorkelund C, Brenner H, Brunner E, Clarke R, Cooper JA, Cremer P, Cushman M, Dagenais GR, D Agostino RB, Sr., Dankner R, Davey-Smith G, Deeg D, Dekker JM, Engstrom G, Folsom AR, Fowkes FG, Gallacher J, Gaziano JM, Giampaoli S, Gillum RF, Hofman A, Howard BV, Ingelsson E, Iso H, Jorgensen T, Kiechl S, Kitamura A, Kiyohara Y, Koenig W, Kromhout D, Kuller LH, Lawlor DA, Meade TW, Nissinen A, Nordestgaard BG, Onat A, Panagiotakos DB, Psaty BM, Rodriguez B, Rosengren A, Salomaa V, Kauhanen J, Salonen JT, Shaffer JA, Shea S, Ford I, Stehouwer CD, Strandberg TE, Tipping RW, Tosetto A, Wassertheil-Smoller S, Wennberg P, Westendorp RG, Whincup PH, Wilhelmsen L, Woodward M, Lowe GD, Wareham NJ, Khaw KT, Sattar N, Packard CJ, Gudnason V, Ridker PM, Pepys MB, Thompson SG, Danesh J. C-reactive protein, fibrinogen, and cardiovascular disease prediction. N Engl J Med. 2012;367: Ridker PM, Kastelein JJ, Genest J, Koenig W. C-reactive protein and cholesterol are equally strong predictors of cardiovascular risk and both are important for quality clinical care. Eur Heart J. 2013;34: Nissen SE, Tuzcu EM, Schoenhagen P, Crowe T, Sasiela WJ, Tsai J, Orazem J, Magorien RD, O Shaughnessy C, Ganz P; Reversal of Atherosclerosis with Aggressive Lipid Lowering (REVERSAL) Investigators. Statin therapy, LDL cholesterol, C-reactive protein, and coronary artery disease. N Engl J Med. 2005;352: Ridker PM, Cannon CP, Morrow D, Rifai N, Rose LM, McCabe CH, Pfeffer MA, Braunwald E; Pravastatin or Atorvastatin Evaluation and Infection Therapy-Thrombolysis in Myocardial Infarction 22 (PROVE IT-TIMI 22) Investigators. C-reactive protein levels and outcomes after statin therapy. N Engl J Med. 2005;352: Kinlay S. Role of C-reactive protein when prescribing a statin. Curr Atheroscler Rep. 2012;14: Nicholls SJ, Ballantyne CM, Barter PJ, Chapman MJ, Erbel RM, Libby P, Raichlen JS, Uno K, Borgman M, Wolski K, Nissen SE. Effect of two intensive statin regimens on progression of coronary disease. N Engl J Med. 2011;365: Nicholls SJ, Borgman M, Nissen SE, Raichlen JS, Ballantyne C, Barter P, Chapman MJ, Erbel R, Libby P. Impact of statins on progression of atherosclerosis: rationale and design of SATURN (Study of Coronary Atheroma by InTravascular Ultrasound: effect of Rosuvastatin versus AtorvastatiN). Curr Med Res Opin. 2011;27: Nicholls SJ, Hsu A, Wolski K, Hu B, Bayturan O, Lavoie A, Uno K, Tuzcu EM, Nissen SE. Intravascular ultrasound-derived measures of coronary atherosclerotic plaque burden and clinical outcome. J Am Coll Cardiol. 2010;55: Baigent C, Blackwell L, Emberson J, Holland LE, Reith C, Bhala N, Peto R, Barnes EH, Keech A, Simes J, Collins R. Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet. 2010;376: Morrow DA, de Lemos JA, Sabatine MS, Wiviott SD, Blazing MA, Shui A, Rifai N, Califf RM, Braunwald E. Clinical relevance of C-reactive protein during follow-up of patients with acute coronary syndromes in the Aggrastat-to-Zocor Trial. Circulation. 2006;114: Ridker PM, Danielson E, Fonseca FA, Genest J, Gotto AM Jr, Kastelein JJ, Koenig W, Libby P, Lorenzatti AJ, MacFadyen JG, Nordestgaard BG, Shepherd J, Willerson JT, Glynn RJ; JUPITER Study Group. Rosuvastatin to prevent vascular events in men and women with elevated C-reactive protein. N Engl J Med. 2008;359: Ridker PM, Danielson E, Fonseca FA, Genest J, Gotto AM Jr, Kastelein JJ, Koenig W, Libby P, Lorenzatti AJ, Macfadyen JG, Nordestgaard BG, Shepherd J, Willerson JT, Glynn RJ; JUPITER Trial Study Group. Reduction in C-reactive protein and LDL cholesterol and cardiovascular event rates after initiation of rosuvastatin: a prospective study of the JUPITER trial. Lancet. 2009;373: Nissen SE, Tuzcu EM, Schoenhagen P, Brown BG, Ganz P, Vogel RA, Crowe T, Howard G, Cooper CJ, Brodie B, Grines CL, DeMaria AN; REVERSAL Investigators. Effect of intensive compared with moderate lipid-lowering therapy on progression of coronary atherosclerosis: a randomized controlled trial. JAMA. 2004;291: Kinlay S. Low-density lipoprotein-dependent and -independent effects of cholesterol-lowering therapies on C-reactive protein: a meta-analysis. J Am Coll Cardiol. 2007;49: Lewis SJ, Sacks FM, Mitchell JS, East C, Glasser S, Kell S, Letterer R, Limacher M, Moye LA, Rouleau JL, Pfeffer MA, Braunwald E. Effect of pravastatin on cardiovascular events in women after myocardial infarction: the cholesterol and recurrent events (CARE) trial. J Am Coll Cardiol. 1998;32: Cannon CP, Braunwald E, McCabe CH, Rader DJ, Rouleau JL, Belder R, Joyal SV, Hill KA, Pfeffer MA, Skene AM; Pravastatin or Atorvastatin Evaluation and Infection Therapy-Thrombolysis in Myocardial Infarction 22 Investigators. Intensive versus moderate lipid lowering with statins after acute coronary syndromes. N Engl J Med. 2004;350: Plenge JK, Hernandez TL, Weil KM, Poirier P, Grunwald GK, Marcovina SM, Eckel RH. Simvastatin lowers C-reactive protein within 14 days: an effect

9 Puri et al Statins, CRP, Plaque Progression, and MACE 2403 independent of low-density lipoprotein cholesterol reduction. Circulation. 2002;106: Wensley F, Gao P, Burgess S, Kaptoge S, Di Angelantonio E, Shah T, Engert JC, Clarke R, Davey-Smith G, Nordestgaard BG, Saleheen D, Samani NJ, Sandhu M, Anand S, Pepys MB, Smeeth L, Whittaker J, Casas JP, Thompson SG, Hingorani AD, Danesh J. Association between C-reactive protein and coronary heart disease: Mendelian randomisation analysis based on individual participant data. BMJ. 2011;342:d Smeeth L, Thomas SL, Hall AJ, Hubbard R, Farrington P, Vallance P. Risk of myocardial infarction and stroke after acute infection or vaccination. N Engl J Med. 2004;351: Ridker PM, Howard CP, Walter V, Everett B, Libby P, Hensen J, Thuren T. Effects of interleukin-1β inhibition with canakinumab on hemoglobin A1c, lipids, C-reactive protein, interleukin-6, and fibrinogen: a phase iib randomized, placebo-controlled trial. Circulation. 2012;126: Clinical Perspective Although inflammation appears to be involved in all stages of atherosclerosis, the mechanistic and prognostic role of inflammation in individuals treated with potent statin therapy remains unclear. The Study of Coronary Atheroma by Intravascular Ultrasound: Effect of Rosuvastatin Versus Atorvastatin (SATURN) was the largest serial atherosclerosis imaging study undertaken in >1000 individuals treated with the highest approved doses of 2 of the most potent statins currently available. We thus undertook a prespecified post hoc analysis to determine the prognostic role of changes in C-reactive protein (CRP) in individuals prescribed 2 years of maximally intensive statin therapy, who achieved a mean on-treatment low-density lipoprotein cholesterol of 65.9 mg/dl. Despite potent statin therapy, over one-third of individuals experienced rising ontreatment CRP levels, with a poor correlation between changes in low-density lipoprotein cholesterol and changes in CRP. Decreasing CRP levels independently associated with coronary atheroma regression. Although the absolute change in CRP was nonprognostic, the on-treatment level of CRP independently associated with major adverse cardiovascular events. A stepwise relationship was noted between on-treatment CRP levels and time to first major adverse cardiovascular event. Collectively, these results highlight the potential importance of inflammation mediating residual cardiovascular risk in individuals with coronary artery disease treated according to current lipid-lowering guidelines. Such findings also underscore the importance of clinical trials that will specifically test the inflammatory hypothesis of atherothrombosis.