A Tale of Two Trials
A Comparison of the Post–Acute Coronary Syndrome Lipid-Lowering Trials A to Z and PROVE IT–TIMI 22
Background— The Aggrastat to Zocor (A to Z) and Pravastatin or Atorvastatin Evaluation and Infection Therapy (PROVE IT) trials compared intensive and moderate statin therapy after acute coronary syndromes, with seemingly disparate results. We analyzed the design, implementation, and results of the two trials in an attempt to clarify the effects of early intensive statin therapy.
Methods and Results— Study design, end points, and definitions were compared. In each trial, comparisons were made between intensive and moderate arms for both trials’ primary end points and death/myocardial infarction. Analyses were performed over various time points: at the end of the trials, ≤4 months, and >4 months. Subjects in A to Z had higher-risk demographics. More PROVE IT subjects were enrolled in the United States and underwent prerandomization revascularization. The low-density lipoprotein (LDL) difference was greater in A to Z than in PROVE IT early (≤4 months) but less late. Significant C-reactive protein reduction was earlier in PROVE IT. With common end points, event rates were higher in A to Z, and early favorable separation of event curves was seen in PROVE IT but not in A to Z. Clinical end point rates and reductions were similar in both trials after 4 months.
Conclusions— An early benefit was seen in PROVE IT but not in A to Z. Late-phase results were similar. Factors that may explain this disparity include the intensity of therapy in the early phase, timing, and magnitude of LDL and C-reactive protein lowering, differences in early revascularization, and the play of chance. Taken together, the results of these trials support a strategy of early intensive statin therapy coupled with revascularization when appropriate in patients after acute coronary syndrome.
Received August 31, 2005; revision received December 23, 2005; accepted January 3, 2006.
Statins, or HMG-CoA reductase inhibitors, have been shown to reduce cardiovascular events in subjects with or at risk for coronary artery disease.1 Clinical benefit in trials with progressively lower baseline and achieved cholesterol levels has led to expanded indications for these medications.2,3 Despite robust and consistent results, questions remained about early initiation of statin therapy after acute coronary syndrome (ACS) and whether intensive therapy would result in a greater benefit.
Editorial p 1382
Clinical Perspective p 1414
Two trials were conducted to address these questions: the Aggrastat to Zocor (A to Z) trial4 and the Pravastatin or Atorvastatin Evaluation and Infection Therapy (PROVE IT)–TIMI 22 trial.5 A to Z was a 2-phase study performed in patients with ST-elevation and non–ST-elevation ACS. The Aggrastat phase, or A-phase, examined the safety and efficacy of enoxaparin versus unfractionated heparin with tirofiban. The Zocor phase, or Z-phase, compared an early/intensive with a delayed/conservative statin therapy strategy.4 PROVE IT examined the safety and efficacy of intensive versus moderate statin therapy.5 A to Z showed a favorable trend in the primary end point for intensive statin therapy that did not achieve statistical significance,6 whereas PROVE IT demonstrated a significant benefit for intensive compared with moderate statin therapy.7
This disparity has led to speculation, especially among clinical trialists, about potential differences between these studies8 and clinical questions of whether different statin therapies have consistent effects. In the present study, we analyze the similarities and differences in trial design, implementation, and results of A to Z and PROVE IT in an attempt to clarify the risks and benefits of early intensive statin therapy after ACS.
The trial designs and treatment regimens used in A to Z and PROVE IT have been previously published.4,5 Both were post-ACS trials; inclusion and exclusion criteria were similar. A to Z subjects were enrolled within 5 days of the presenting event and required “clinical stability” for 12 hours, whereas PROVE IT subjects could be enrolled within 10 days and required “clinical stability” for 24 hours; only A to Z excluded prior statin therapy.4,5
The treatment regimens are summarized in Figure 1. In A to Z, the early/intensive therapy arm received 40 mg simvastatin for 1 month and then 80 mg for the remainder of the trial; the delayed/conservative (moderate) therapy arm received placebo for 4 months followed by 20 mg simvastatin for the remainder of the trial. In PROVE IT, subjects received either 80 mg atorvastatin (intensive) or 40 mg pravastatin (moderate) for the entire trial. Lipid levels and high-sensitivity (hs)–C-reactive protein (CRP) were measured in central laboratories for both trials.
The Z-phase of A to Z was designed to continue until 970 subjects experienced the primary end point,4 providing 90% power to detect a 20% reduction. Although the end point rate was lower than modeled, the trial’s Executive Committee elected to halt enrollment at 4500 subjects, in part because of emerging data on the early efficacy of statins after ACS9 and increasing difficulty enrolling subjects into a placebo-controlled trial due to changing practice patterns. By trial completion, 652 subjects had primary end points, reducing statistical power.
PROVE IT was designed to assess clinical noninferiority defined by the upper confidence limit of relative risk between the two agents being no greater than 1.17.5 The trial was planned to continue until 925 subjects had reached the primary end point. By the trial conclusion, 1001 subjects had a primary end point, increasing statistical power.
For A to Z, the primary composite end point was cardiovascular death, nonfatal myocardial infarction (MI), readmission for ACS, and stroke. The PROVE IT primary composite end point was all-cause death, MI, unstable angina requiring hospitalization, revascularization (>30 days after randomization), and stroke. There were subtle differences in MI and ischemic end point definitions.4,5 In both trials, major efficacy end points were adjudicated by independent clinical events committees.
The authors had full access to the data and take full responsibility for its integrity. All authors have read and agree to the manuscript as written. For this manuscript, efficacy end points were determined from patient-level full-study databases. Readmission for ACS (A to Z) and unstable angina (PROVE IT) were considered equivalent as were the definitions of MI. In each trial, comparisons were made between intensive and moderate therapy arms for the A to Z and PROVE IT primary end points and death/MI. Analyses were performed over various time points: at the end of the trials, <4 months (duration of placebo therapy in A to Z), and >4 months. “Death” or “mortality” refers to all-cause death unless otherwise noted. Analyses of the time period >4 months are conditional analyses that exclude patients with the end point in question during the first 4 months.
For the purpose of the present analysis, efficacy analyses were performed with survival analysis methods. Full-study results are presented as 2-year Kaplan-Meier estimates (730 days) to ensure comparable data between trials. For A to Z, the main analyses for the primary manuscript were performed at 790 days.4,6 Main analyses in PROVE IT and A to Z adjusted for different baseline features (factorial gatifloxacin assignment in PROVE IT and age in A to Z).4–7 To provide directly comparable analyses, no such adjustments were made in the present study. As a result of these analytical differences, event rates and statistical measures vary slightly from the main trial manuscripts.6,7 Statistical comparisons were made within each trial by using all available data through the end of follow-up. Because of differences in adverse event collection, including more frequent monitoring leading to downtitrations and discontinuation of study drug in PROVE IT, differences in definitions of adverse events and methods of reporting4,5 a direct comparison of safety events is problematic. Therefore, results presented herein are limited to published rates of muscle and liver enzyme elevations, and supportive analyses of muscle enzyme elevations in PROVE IT were performed to provide comparison where published information was not available. Efficacy results are expressed as hazard ratios; 2-sided P values <0.05 are considered significant.
To compare end points between trials, regression models were created for each end point and time point, and tests for interaction between trials were included. Trial results were considered to be different if the interaction probability value was significant at the P<0.05 level. Individual patients were pooled, and separate multivariable regression models were constructed for death/MI and the A to Z and PROVE IT end points. Multivariable models were constructed by using backward selection of univariate predictors of outcome including age, gender, index diagnosis, diabetes, prior MI, smoking, and percutaneous coronary intervention (PCI). Variables for intensive therapy, trial, and enrollment within or outside of the United States were added.
Within each trial, baseline characteristics in each treatment arm were well matched and are presented by trial (Table 1). Subjects in A to Z were older, more frequently diabetic, smokers, and less frequently white. No subject in A to Z had prior statin therapy compared with 25% of subjects in PROVE IT. Subjects in A to Z were enrolled significantly earlier than in PROVE IT. Both trials had high rates of use of evidence-based medications for treatment of ACS.6,7
There was a marked difference between the two trials in the geographic distribution of enrollment (Table 1); 21% of subjects in A to Z were enrolled in the United States, compared with 71% in PROVE IT. More subjects in PROVE IT underwent PCI for the index event. Region of enrollment and rate of PCI were closely related; 79% of patients in each trial enrolled in the United States underwent prerandomization PCI.
Lipid and hs-CRP Results
Baseline lipids were similar between the two trial populations (Table 1). The subjects in A to Z had slightly higher baseline total and low-density lipoprotein (LDL) cholesterol values and lower triglycerides.
Median LDL values were reduced in all active treatment groups (Figure 2A). At 1 and 4 months, there was a greater difference between the two strategies in A to Z compared with PROVE IT, largely related to higher LDL in the moderate (placebo) arm. By 8 months, the intensive therapy arms remained similar, but the A to Z delayed/conservative arm (20 mg simvastatin) had lower LDL than the PROVE IT moderate therapy arm (40 mg pravastatin). Consequently, after 4 months (initiation of 20 mg simvastatin), the difference in median LDL between treatment arms was less in A to Z than in PROVE IT (Figure 2A).
During treatment, high-density lipoprotein (HDL) was higher than baseline in all treatment groups. At 4 months in A to Z, the intensive therapy arm had higher levels of HDL than the moderate therapy arm (43 versus 41, P<0.0001), whereas in PROVE IT, the intensive therapy arm had lower levels of HDL (40 versus 42, P<0.0001) than the moderate therapy arm.
At baseline, median CRP levels were higher in A to Z (20 mg/L) than in PROVE IT (12 mg/L), consistent with the earlier enrollment of subjects after ACS. By 1 month, levels had decreased substantially in both trials (Figure 2B). The intensive therapy arm in A to Z (40 mg simvastatin) showed a small, nonsignificant difference in CRP compared with placebo, whereas median CRP level was significantly lower in the intensive therapy arm in PROVE IT. By 4 months, there was continued lowering of CRP in the intensive treatment arms of both trials, yielding significant differences. A greater proportion of the intensively treated subjects in PROVE IT reached dual hs-CRP value (<2 mg/L) and LDL (<70 mg/dL) cutoffs10,11 at 1 month than in A to Z (44% versus 24%), though this difference narrowed by 4 months (41 versus 36%).
Overall Trial Efficacy Results
Overall event rates were higher in A to Z than in PROVE IT (Table 2). Using the A to Z composite end point, the A to Z intensive therapy arm showed a hazard ratio (HR) of 0.89 (P=0.12). When the same end point was applied to the PROVE IT database, the intensive therapy arm showed a HR of 0.82 (P=0.03). Pinteraction=0.52 for difference between trials. When the PROVE IT composite end point was applied to the A to Z database, the intensive therapy arm showed a HR of 1.0 (P=0.79), whereas in the PROVE IT trial the intensive therapy arm showed a HR of 0.84 (P=0.005). This difference between trials was statistically significant (Pinteraction=0.03). The composite of death/MI showed results similar to the A to Z primary end point, though with fewer events, and did not reach formal statistical significance in PROVE IT.
Timing of Events
Regardless of the end point chosen, there were higher rates of end points in A to Z than in PROVE IT in the early phase (Table 2). No benefit was observed for intensive therapy within the first 4 months in A to Z (Table 2, Figure 3A). However, in PROVE IT, a nearly significant (using the A to Z end point, death/MI) or significant (using the PROVE IT end point) reduction was observed within 4 months (Table 2, Figure 3B). Using the PROVE IT end point, the difference between the two trials was statistically significant (Pinteraction=0.003) at this time point, whereas for death/MI (Pinteraction=0.10) and the A to Z composite (Pinteraction=0.14), the trends toward greater benefit in PROVE IT were not significant.
When a conditional analysis was performed, censoring subjects with events before 4 months, subsequent event rates and the effects of intensive statin therapy compared with moderate statin therapy were more similar (Table 2), with HR of 0.83 (PROVE IT) and 0.76 (A to Z) using the A to Z end point and 0.86 HR in both trials using the PROVE IT end point. There was no statistical difference (interaction) between the trials after 4 months.
In A to Z, alanine aminotransferase (ALT) elevation >3 times the upper limit of normal (ULN) was observed in 0.4% of patients receiving moderate therapy compared with 0.9% of patients receiving intensive statin therapy (P=0.05).6 In PROVE IT, this level of ALT elevation was detected in 1.1% versus 3.3% of patients (P<0.001).7 Creatine kinase (CK) elevation greater than 10 times ULN was observed in 0.04%, and 0.4% in the moderate and intensive therapy arms of A to Z; 3 subjects in the intensive therapy arm had CK >10 000 IU.6 CK elevation >10 times ULN was observed in 0.1% and 0.15% in the moderate and intensive therapy arms of PROVE IT; no subject had CK >10 000 IU.
Site of Enrollment and Revascularization
In each trial, the relative benefit of intensive statin therapy tended to be greater among subjects enrolled in the United States (Table 3). In A to Z, there were benefits seen in the late phase (>4 months), favoring intensive therapy. In PROVE IT, there were consistent benefits across the time periods among subjects enrolled in the United States. In contrast, no significant benefit was seen at any time point within either trial among subjects enrolled outside of the United States. In PROVE IT, early benefit was observed among patients enrolled in the United States but not those outside the United States. Regardless of region of enrollment, there was no early benefit in A to Z (Table 3). The most significant difference between patients enrolled in the United States and outside of the United States was the frequency of revascularization for the index ACS event; 79% versus 34% (P<0.001) in A to Z and 79% versus 44% in PROVE IT (P<0.001). When stratified for revascularization for the index event, the greatest benefit was seen in patients who underwent revascularization (Table 4). Greater trends toward benefit (<4 months) were seen in PROVE IT patients who had early revascularization than in those who did not. No early benefit was seen in A to Z regardless of whether or not patients were revascularized.
When data were pooled between trials, baseline features that were predictors of death/MI during follow-up were age, diabetes mellitus, cigarette smoking, and history of MI (Table I in the online Data Supplement). Subjects who had a PCI for the qualifying event and those who were allocated to intensive treatment arms had lower rates of death/MI during follow-up. When accounting for these features, non-US randomization did not remain independently predictive of death/MI (HR =0.97, P=0.73). Multivariate predictors of the primary composite end points were largely similar (Tables II and III in the online Data Supplement), with US enrollment only associated with outcomes for the PROVE IT end point.
Until recently, few data were available on the optimal timing and intensity of initiation of statin therapy after ACS. The A to Z and PROVE IT trials each examined intensive versus moderate statin therapy after ACS with seemingly disparate results and were analyzed in the present study using primary data. The major findings from these analyses suggest that the near totality of the differences between the results of these two trials is explained by PROVE IT’s, but not A to Z’s, showing an early benefit for intensive therapy. Several features including intensity of early statin dose, rapidity and extent of CRP lowering, and disparities in early use of revascularization procedures seem to account, at least in part, for this difference in early effect.
In the first 4 months of PROVE IT but not in A to Z, an event reduction was seen with intensive statin therapy. Several possibilities exist for the differences in the early phase. Subjects in A to Z had more baseline risk factors that increase the risk of early events. In addition, subjects in A to Z were enrolled earlier than those in PROVE IT (mean, 3.7 versus 5.7 days), with less time to achieve clinical stability. Fewer subjects in A to Z compared with those in PROVE IT had revascularization for the index presentation and therefore received less definitive therapy for the culprit artery involved in the index event. These factors may have contributed to the higher event rate (especially early events) in A to Z compared with PROVE IT. This hypothesis is supported by the benefit seen in PROVE IT patients who had prerandomization PCI and absence of benefit in those who did not. In nonrevascularized patients, the dominant pathophysiological determinant of early clinical events may be thrombosis associated with the culprit lesion and may be less modifiable by intensive statin treatment.
Extent of LDL lowering seems unlikely to be an important contributor to the early differences in these trials. Previous studies suggest a direct relation between LDL cholesterol lowering and clinical benefits.1 The relative LDL difference was greater in A to Z than PROVE IT in the first 4 months. However, in the later phases of the trials, the relative LDL difference was greater in PROVE IT. It is possible that LDL lowering may be more closely related to long-term than to short-term outcomes.12
It could therefore be reasoned that LDL reduction alone may not have been sufficient to result in an early benefit and raises the question of a noncholesterol (pleiotropic) effect of statin therapy, such as antiinflammatory or antithrombotic effects,13,14 accounting in part for the differences in timing and magnitude of benefit. By 1 month, PROVE IT demonstrated a significant difference in CRP between treatment groups, whereas A to Z did not. By 4 months, both trials showed significant differences in CRP. This pattern of early benefit with intensive statin therapy was previously observed in the Myocardial Ischemia Reduction With Aggressive Cholesterol Lowering (MIRACL) trial, which enrolled subjects after ACS but excluded an early invasive strategy.9 Early event rates in MIRACL were higher than those seen in A to Z and PROVE IT, and, despite the noninvasive strategy, an early separation of event curves was seen. This early event reduction also coincided with a significant 16-week reduction in CRP.15
Recent analyses have shown an additive benefit for CRP and LDL reduction,10,11 with greatest benefits occurring in subjects who had lowering of LDL and CRP to predefined targets. These “dual targets” were more frequently achieved at 1 month in the subjects treated with 80 mg of atorvastatin in PROVE IT than in those treated with 40 mg of simvastatin in A to Z and coincident with the early clinical benefit. By 4 months (with 80 mg simvastatin and 80 mg atorvastatin), however, the rates of achievement of these dual targets were more similar, with similar clinical outcomes in the two trials. It is not possible to determine if a similar early separation of event curves would have been seen if the initial therapy in A to Z had been 80 mg of simvastatin, the highest dose in the trial, or if the differences in results are intrinsic to the two statins.
The similarity of the benefit in the two trials in the late phase (>4 months) is striking. When using common end points, the rates of events and event reductions were very similar during the late phase. These findings suggest that in the late phase, intensive statin therapy continues to provide benefit over moderate-intensity therapy. We hypothesize that by 4 months, subjects are beyond the early “thrombotic/inflammatory phase” and are more stable than immediately after ACS. Thus, the differences between the trials in enrollment time, site, and presence of early revascularizations would play a lesser role in outcomes.
A recently published meta-analysis suggests that clinical benefit of statin versus placebo is proportional to on treatment change in LDL.1 This holds true for the 2-year results of A to Z and PROVE IT, with the greater clinical benefit occurring PROVE IT with greater long-term LDL differences. However, this does not explain the early differences, and although the late-phase LDL differential was greater in PROVE IT than A to Z, clinical benefits were similar in both trials. One possible explanation is the relatively greater increase in HDL cholesterol in patients treated with high-dose simvastatin compared with those treated with high-dose atorvastatin, a finding consistent with previous studies.16,17
Patients who have survived 4 months after ACS without a recurrent event could be considered to be patients with “stable coronary artery disease (CAD).” Data from the Treating to New Targets (TNT) and the Incremental Decrease in End points through Aggressive Lipid lowering (IDEAL) trials are consistent with the findings18,19 in the chronic phases of A to Z and PROVE IT.
In clinical practice, the benefits of any therapy must be balanced against risk. Major statin side effects are known to be agent- and dose-related.20 In both A to Z and PROVE IT, intensive therapies were well tolerated. Documented muscle enzyme elevations were more frequent with high-dose simvastatin in A to Z, whereas liver enzyme elevations were more frequent with high-dose atorvastatin in PROVE IT. Patients at highest risk of adverse effects are often excluded from clinical trials. Therefore, rates of side effects among ACS patients in clinical practice may be higher than presented herein (and in other trials), and it is important for the clinician to monitor patients closely for side effects and to discontinue or dose-reduce statin therapy when important side effects occur.
Important limitations in an analysis of two separate trials must be noted. Though analyses were performed on the full databases and definitions of end points were similar, the processes of end point collection and adjudication differed. Because of the different timing of therapy initiation and titration steps in A to Z, it is difficult to make direct comparisons between the drugs used in the intensive or moderate therapy arms between trials. All comparisons between trials must be considered hypothesis-generating, as they are post hoc. Attempts to perform analyses stratified by specific variables in the trial such as country of origin and revascularization status are limited by strong links between trial enrollment and patient demographics.
Both the A to Z and PROVE IT trials showed numerically lower rates of cardiovascular events over a 2-year period with more intensive, compared with less intensive, statin therapy. Overall, the reduction was greater and achieved statistical significance in PROVE IT, whereas A to Z trended toward benefit. Late-phase results were similar, showing similar benefits for more intensive versus less intensive statin therapy. An early benefit for intensive statin therapy was seen in PROVE IT but not in A to Z. Factors that may explain these discrepancies include differences in the baseline risk factors, rates of early revascularization, intensity of early therapy, timing and magnitude of LDL and CRP lowering, and the play of chance.
Although revascularization was not randomized, these data support a strategy of early intensive statin therapy coupled with revascularization (in appropriate candidates) after ACS. The relation of early benefit to reductions of CRP suggests that inflammation may play an important role in early events. Strategies targeting both early aggressive LDL and CRP lowering deserve further prospective study.
The A to Z and PROVE IT trials compared intensive and moderate statin therapy after ACS, with seemingly disparate results. The present study analyzed the design, implementation, and results of the two trials by using similar end points, in an attempt to clarify the effects of early intensive statin therapy. Major differences between the trials included higher-risk subjects enrolled in A to Z and more subjects enrolled in PROVE IT who had undergone revascularization. In the intensive therapy arms, the early-phase treatment in PROVE IT was more potent than in A to Z (80 mg atorvastatin versus 40 mg simvastatin), whereas late phase treatment was more similar (80 mg atorvastatin versus 80 mg simvastatin). More subjects had clinical events in A to Z, overall. A benefit was seen early (within 4 months) in PROVE IT, and no early benefit was seen in A to Z. In both trials, the long-term benefit was similar. Taken together, the results of these trials support a strategy of early intensive statin therapy coupled with revascularization when appropriate in patients after ACS.
Merck Inc provided support for the A to Z Trial, and Bristol Myers Squibb provided support for the PROVE IT–TIMI 22 Trial.
The TIMI study group receives major research grant support from Bristol Myers Squibb, Merck and Co, and Pfizer, Inc (Drs Stephen D. Wiviott, Sabina A. Murphy, Carolyn H. McCabe, Eugene Brunwald). Dr Stephen D. Wiviott has served on the Speakers’ Bureau for Pfizer (minor) and received honoraria from Merck (minor). Dr James A. de Lemos has received research grant support from Merck and Co (minor) and has served on speakers’ bureaus for Merck, Merck-Schering Plough, and Pfizer (minor). Dr Christopher P. Cannon has received major research grant support from Bristol Myers Squibb, Merck, Merck/Schering-Plough, and Pfizer; has received minor honoraria from Merck, Pfizer, and Schering-Plough; and has served on advisory boards for Bristol Myers-Squibb, Merck, Merck/Schering-Plough, and Pfizer. Duke Clinical Research Institute (Dr Michael Blazing) has received research grant support from Merck (major) and Merck-Schering Plough (major), and Dr Blazing has served on speakers bureaus for Pfizer (minor) and Merck (minor). Dr Robert Califf has received major research grant support from and has had institutional relationships with Merck, Schering-Plough, Pfizer, Novartis, and BMS.
Baigent C, Keech A, Kearney PM, Blackwell L, Buck G, Pollicino C, Kirby A, Sourjina T, Peto R, Collins R, Simes R. Efficacy and safety of cholesterol-lowering treatment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet. 2005; 366: 1267–1278.
Blazing MA, De Lemos JA, Dyke CK, Califf RM, Bilheimer D, Braunwald E. The A-to-Z Trial: methods and rationale for a single trial investigating combined use of low-molecular-weight heparin with the glycoprotein IIb/IIIa inhibitor tirofiban and defining the efficacy of early aggressive simvastatin therapy. Am Heart J. 2001; 142: 211–217.
de Lemos JA, Blazing MA, Wiviott SD, Lewis EF, Fox KA, White HD, Rouleau JL, Pedersen TR, Gardner LH, Mukherjee R, Ramsey KE, Palmisano J, Bilheimer DW, Pfeffer MA, Califf RM, Braunwald E. Early intensive vs a delayed conservative simvastatin strategy in patients with acute coronary syndromes: phase Z of the A to Z trial. JAMA. 2004; 292: 1307–1316.
Law MR, Wald NJ, Thompson SG. By how much and how quickly does reduction in serum cholesterol concentration lower risk of ischaemic heart disease? BMJ. 1994; 308: 367–372.
Kinlay S, Schwartz GG, Olsson AG, Rifai N, Leslie SJ, Sasiela WJ, Szarek M, Libby P, Ganz P. High-dose atorvastatin enhances the decline in inflammatory markers in patients with acute coronary syndromes in the MIRACL study. Circulation. 2003; 108: 1560–1566.
Davidson MH, Ose L, Frohlich J, Scott RS, Dujovne CA, Escobar ID, Bertolami MC, Cihon F, Maccubbin DL, Mercuri M. Differential effects of simvastatin and atorvastatin on high-density lipoprotein cholesterol and apolipoprotein A-I are consistent across hypercholesterolemic patient subgroups. Clin Cardiol. 2003; 26: 509–514.
Illingworth DR, Crouse JR III, Hunninghake DB, Davidson MH, Escobar ID, Stalenhoef AF, Paragh G, Ma PT, Liu M, Melino MR, O’Grady L, Mercuri M, Mitchel YB. A comparison of simvastatin and atorvastatin up to maximal recommended doses in a large multicenter randomized clinical trial. Curr Med Res Opin. 2001; 17: 43–50.
Pedersen TR, Faergeman O, Kastelein JJ, Olsson AG, Tikkanen MJ, Holme I, Larsen ML, Bendiksen FS, Lindahl C, Szarek M, Tsai J. High-dose atorvastatin vs usual-dose simvastatin for secondary prevention after myocardial infarction: the IDEAL study: a randomized controlled trial. JAMA. 2005; 294: 2437–2445.
Pasternak RC, Smith SC Jr, Bairey-Merz CN, Grundy SM, Cleeman JI, Lenfant C. ACC/AHA/NHLBI Clinical Advisory on the Use and Safety of Statins. Stroke. 2002; 33: 2337–2341.
The Aggrastat to Zocor (A to Z) and the Pravastatin or Atorvastatin Evaluation and Infection Therapy (PROVE IT) trials compared intensive and moderate statin therapy after acute coronary syndrome (ACS), with seemingly disparate results. The present study analyzed the design, implementation, and results of the two trials by using similar end points, in an attempt to clarify the effects of early intensive statin therapy. Major differences between the trials included higher-risk subjects enrolled in A to Z and more subjects enrolled in PROVE IT who had undergone revascularization. In the intensive therapy arms, the early-phase treatment in PROVE IT was more potent than in A to Z (80 mg atorvastatin versus 40 mg simvastatin), whereas late-phase treatment was more similar (80 mg atorvastatin versus 80 mg simvastatin). More subjects had clinical events in A to Z, overall. A benefit was seen early (within 4 months) in PROVE IT, and no early benefit was seen in A to Z. In both trials, the long-term benefit was similar. Taken together, the results of these trials support a strategy of early intensive statin therapy coupled with revascularization when appropriate in patients after ACS.
The online-only Data Supplement can be found at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.105.586347/DC1.
Guest Editor for this article was Raymond J. Gibbons, MD.