Cholesterol Lowering, Cardiovascular Diseases, and the Rosuvastatin-JUPITER Controversy a Critical Reappraisal
Total Page:16
File Type:pdf, Size:1020Kb
ORIGINAL INVESTIGATION Cholesterol Lowering, Cardiovascular Diseases, and the Rosuvastatin-JUPITER Controversy A Critical Reappraisal Michel de Lorgeril, MD; Patricia Salen, BSc; John Abramson, MD; Sylvie Dodin, MD; Tomohito Hamazaki, PhD; Willy Kostucki, MD; Harumi Okuyama, PhD; Bruno Pavy, MD; Mikael Rabaeus, MD Background: Among the recently reported cholesterol- tality from stroke and myocardial infarction. Cardiovas- lowering drug trials, the JUPITER (Justification for the cular mortality was surprisingly low compared with total Use of Statins in Primary Prevention) trial is unique: it mortality—between 5% and 18%—whereas the expected reports a substantial decrease in the risk of cardiovascu- rate would have been close to 40%. Finally, there was a lar diseases among patients without coronary heart dis- very low case-fatality rate of myocardial infarction, far from ease and with normal or low cholesterol levels. the expected number of close to 50%. The possibility that bias entered the trial is particularly concerning because of Methods: Careful review of both results and methods the strong commercial interest in the study. used in the trial and comparison with expected data. Conclusion: The results of the trial do not support the Results: The trial was flawed. It was discontinued (ac- use of statin treatment for primary prevention of cardio- cording to prespecified rules) after fewer than 2 years of vascular diseases and raise troubling questions concern- follow-up, with no differences between the 2 groups on ing the role of commercial sponsors. the most objective criteria. Clinical data showed a major discrepancy between significant reduction of nonfatal stroke and myocardial infarction but no effect on mor- Arch Intern Med. 2010;170(12):1032-1036 HE RESULTS OF RECENT CHO- The JUPITER trial tested the effects of lesterol-lowering drug trials rosuvastatin therapy (20 mg/d) in pa- Author Affiliations: Laboratoire on decreasing morbidity and tients without cardiovascular history or Cœur and Nutrition, Faculty of Medicine, Universite´ Joseph mortality among persons established CHD and with normal or low Fourier and Centre National de with or without coronary cholesterol levels but relatively high lev- la Recherche Scientifique, heart disease (CHD) have been consis- els of C-reactive protein, a fluctuating T 1-9 18 Grenoble, France (Dr de Lorgeril tently negative. However, there is one biologic marker of inflammation. The au- and Ms Salen); Harvard Medical exception, the JUPITER (Justification for thors reported a 50% reduction in low- School, Boston, Massachusetts the Use of Statins in Primary Prevention) density lipoprotein cholesterol levels, a (Dr Abramson); Department of 37% reduction in C-reactive protein lev- Obstetrics and Gynaecology, See also pages 1007, els, and a roughly 50% decrease in car- Laval University, Quebec City, diovascular complications.10 The publica- Quebec, Canada (Dr Dodin); 1024, and 1073 tion of the JUPITER trial (in November Department of Clinical Sciences, 2008) was much anticipated since the Institute of Natural Medicine, trial,10 which showed—in primary preven- University of Toyama, Japan announcement of the trial’s premature dis- tion—a striking decrease in CHD compli- continuation in March 2008,19,20 at a meet- (Dr Hamazaki); Cardiology cations. The JUPITER trial rapidly pro- Department, Clinique Antoine 11-13 ing of the American College of Cardiol- Depage, Brussels, Belgium voked controversy regarding both the ogy, following the presentation of the (Dr Kostucki); Open Research results and the methods used in the trial. disappointing results of the ENHANCE Center for Lipid Nutrition, Kinjo Although enthusiastic comments have (Ezetimibe and Simvastatin in Hypercho- 14-17 Gakuin University, Nagoya, been published, and the results have lesterolemia Enhances Atherosclerosis Japan (Dr Okuyama); undoubtedly propelled many healthy per- Regression) trial.9 Similarly to ezetimibe Re´adaptation Cardiovasculaire, sons without elevated cholesterol levels onto (tested in the ENHANCE trial), rosuva- Centre Hospitalier long-term statin treatment, the clinical rel- de Machecoul, Machecoul, statin was already the subject of aggres- France (Dr Pavy); and evance of the JUPITER trial remains in ques- sive marketing despite the absence of evi- Cardiology Department, tion. To understand the rosuvastatin- dence that its use actually decreased CHD Clinique de Genolier, Genolier, JUPITER controversy, we critically review complications. Indeed, disregarding open- Switzerland (Dr Rabaeus). several significant issues of that study. label studies such as ASTEROID (A Study (REPRINTED) ARCH INTERN MED/ VOL 170 (NO. 12), JUNE 28, 2010 WWW.ARCHINTERNMED.COM 1032 ©2010 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/30/2021 to Evaluate the Effect of Rosuvastatin on Intravascular Ultrasound-Derived Coronary Atheroma Burden),21 3 Table. A Summary of the JUPITER Trial Resultsa trials with rosuvastatin (CORONA [Controlled Rosu- vastatin Multinational Trial in Heart Failure],1 GISSI-HF Rosuvastatin Placebo Group Group [Gruppo Italiano per lo Studio della Sopravvivenza End Point (n=8901) (n=8901) 3 nell’Infarto Miocardio–Heart Failure], and AURORA [A b Study to Evaluate the Use of Rosuvastatin in Subjects on Primary end point 142 251 Nonfatal myocardial infarction 22 62 Regular Haemodialysis: An Assessment of Survival and Any myocardial infarction 31 68 5 Cardiovascular Events] ) had been conducted, and all had Nonfatal stroke 30 58 failed to provide evidence that rosuvastatin therapy re- Any stroke 33 64 duces CHD complications. The failure of rosuvastatin to Arterial revascularization 71 131 show a significant protective effect was also true for pa- Hospitalization for unstable angina 16 27 tients with established CHD, because most patients in the Myocardial infarction, stroke, or confirmed 83 157 deaths from cardiovascular causes CORONA and GISSI-HF trials were survivors of a pre- Death from any cause on known date 190 235 vious myocardial infarction. a Adapted from Ridker et al.10 METHODOLOGICAL PROBLEMS b A combination of myocardial infarction, stroke, arterial revascularization, IN THE JUPITER TRIAL hospitalization for unstable angina, or death from cardiovascular causes. The JUPITER trial was prematurely terminated. Although CLINICAL AND EPIDEMIOLOGICAL having prespecified early stopping rules is a well-accepted INCONSISTENCIES IN THE JUPITER TRIAL feature of clinical trials, it is critical that the rules truly be prespecified. In the case of the JUPITER trial, the prespeci- In any trial, the consistency of clinical data must be ex- fied rules were not detailed in the published description of amined to determine whether methodological flaws have the study protocol.22 Indeed, we still do not know which increased the risk of bias. For instance, in cardiology, com- end point was used to define them, or which level of ben- parison of the rate of hard end points—fatal and nonfa- efits—unexpected on the basis of the a priori calculated hy- tal myocardial infarction and stroke, which represent most pothesis22—was required to justify early termination. Also, cardiovascular complications in any population—to those it was recently shown that truncated trials are associated expected from a comparable population, at least in the with greater effect sizes than trials that are not stopped early, placebo group, provides such a check on methodology. and this effect is independent of the presence of statistical At first glance (Table), the difference between the 2 stopping rules.23 In defending the decision to end the trial groups in terms of hard end points seems impressive (157 early, the JUPITER investigators stated that the decision vs 83 for placebo and rosuvastatin, respectively). But are was not made by them but by members of an independent these differences plausible? Although an “unequivocal re- safety-monitoring board.24 However, the chairman of this duction in cardiovascular mortality” was announced in board—an investigator of the Clinical Trial Service Unit of March 2008 as the main justification for the premature trial Oxford University, Oxford, England—has been, and still termination,19,20 the absence of cardiovascular mortality is, involved in many other industry-sponsored lipid- data in the published article is striking. One may infer from lowering trials, raising issues of conflict of interest.25,26 the Table—although not indicated in the text—that the Fueling concern about the termination of the study is total number of fatal myocardial infarctions was 9 in the that the data are not consistent with a large difference be- rosuvastatin group (the difference between 31 “any myo- tween treatment and placebo. The primary end point cardial infarctions” and 22 “nonfatal myocardial infarc- (Table, line 1) is a composite of cardiovascular compli- tions”) and 6 (68−62) in the placebo group. Similar cal- cations, some of which—such as revascularization and hos- culations for fatal stroke (the difference between “any pital admission—are of less relevance because they are not stroke” and “nonfatal stroke”) show 3 (33−30) in the ro- complications but medical decisions. Taking only the hard suvastatin group and 6 (64−58) in the placebo group. end points of fatal and nonfatal myocardial infarction and Cardiovascular mortality (fatal stroke plus fatal myo- stroke (Table, line 8)—the end points that are less open cardial infarction) would therefore be identical in both to bias and