JUPITER statin trial a biased sham

This paper was published in the Archives of Internal Medicine 2010;170(12):1032-1036

Study title and authors:
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


This paper can be accessed at: http://archinte.ama-assn.org/cgi/content/full/170/12/1032#REF-IOI05093-1


Dr. de Lorgeril notes that the results of cholesterol-lowering drug trials show no evidence that statin drugs lower the disease rates or death rates of people with or without coronary heart disease with one exception, and that is the JUPITER (Justification for the Use of Statins in Primary Prevention) trial. JUPITER reports a substantial decrease in the risk of cardiovascular diseases among patients without coronary heart disease and with normal or low cholesterol levels. 


The results of the JUPITER study were met with a massive media fanfare proclaiming the benefits of statin drugs. This enthusiastic recommendation has no doubt persuaded many people with normal cholesterol levels to start long term statin treatment.


The JUPITER trial tested the effects of rosuvastatin in patients without heart disease and with normal or low cholesterol levels but relatively high levels of C-reactive protein, a marker of inflammation. The study spanned 1,315 sites in 26 countries and included 17,802 people who were assigned either 20 mg/d of rosuvastatin or placebo.


3 recent trials with rosuvastatin (with the acronyms CORONA, GISSI-HF and AURORA) had been conducted, and all had failed to provide evidence that rosuvastatin therapy reduces heart disease complications.


The JUPITER trial was prematurely terminated on the grounds that it had generated evidence that the statin treatment had definitely reduced heart disease rates.


However the evidence shows otherwise:
(a) If you include people who had fatal and nonfatal heart attack and stroke - the trial was stopped after only 240 incidents. 
(b) There was no difference in the incidence of serious adverse events (total hospitalizations, prolongations of hospitalizations, cancer, and permanent disability) between the 2 groups.
(c) There was hardly any difference in death rates when the trial was ended, and the trend was showing that the statin groups death rate was increasing compared to the placebo group.


An "unequivocal reduction in cardiovascular mortality" was announced in March 2008 as the main justification for the premature trial termination.


However the actual facts again beg to differ:
(d) Fatal heart attacks were 9 in the statin group and 6 in the placebo group.
(e) Stroke death was 3 in the statin group compared to 6 taking the placebo.
So there was 12 cardiovascular deaths in each group. Hardly an "unequivocal reduction in cardiovascular mortality" as the JUPITER study authors concluded.


So why was the trial stopped early?


As stated earlier JUPITER was hailed in the media as a ringing endorsement for us all to start statin therapy. This was achieved by the authors of the study only highlighting some results of the trial and completely ignoring other, less favourable data. It also raises the suspicion that if the trial had continued then the results would have shown statins in an even more unfavourable light.


Rosuvastatin (sold under the brand name Crestor) is marketed and distributed by AstraZeneca Pharmaceuticals.


The JUPITER trial involved multiple conflicts of interest: 
(f) It was conducted by Astra Zeneca  with their obvious commercial interests. 
(g) Nine of 14 authors of the JUPITER article have financial ties to the Astra Zeneca. 
(h) The principal investigator has a personal conflict of interest as a co-holder of the patent for the C-reactive protein test.
(i) Astra Zenecas own investigators controlled and managed the raw data which increases the chance of bias appearing in the data.


Dr. de Lorgeril concludes:
(i) The results of the JUPITER trial are clinically inconsistent and therefore should not influence medical practice or clinical guidelines. 
(ii) The results of the JUPITER trial show that commercially sponsored clinical trials are at risk of poor quality and bias. 
(iii) The failure of the JUPITER trial to demonstrate a protective effect of rosuvastatin confirms the results of  more than 12 other cholesterol-lowering trials published in recent years, which all provided no evidence of protection against heart disease by cholesterol lowering. 
(iv) These failed trials strongly suggest that the presumed preventive effects of cholesterol-lowering drugs have been considerably exaggerated.


Dr.de Lorgeril ends by saying that the time has come for a critical reappraisal of cholesterol-lowering and statin treatments for the prevention of heart disease, and the emphasis on pharmaceuticals for the prevention of heart disease has diverted individual and public health attention away from other proven methods of prevention such as a healthy lifestyle, exercise and diet.