JUPITER Analysis Helps Assuage Fears About Diabetes Risk With Statins

August 13, 2012

August 9, 2012 (Boston, Massachusetts) — In the JUPITER primary-prevention trial, the risk of developing diabetes mellitus with statin therapy is limited to patients already at a high risk for developing diabetes, such as those with impaired fasting glucose, metabolic syndrome, severe obesity, or raised hemoglobin A1c (HbA1c) levels [1]. However, in these high-risk patients, as well as in the entire study cohort, the benefits of statin therapy exceeded the risk of diabetes and should reassure physicians about the use of statins for the primary prevention of reducing MI, stroke, and cardiovascular death, say researchers.

The study is published in the August 11, 2012 issue of the Lancet.

"This is a really important clinical issue to get right, particularly because of media coverage earlier this year when concern was raised that patients taking statins had an increased risk of developing diabetes, and on that basis many patients stopped taking their medications," lead investigator Dr Paul Ridker (Brigham and Women's Hospital, Boston, MA) told heartwire . "Unfortunately, little if any data were available at that time to address not only the risks but also the benefits of treatment. This is crucial, since it is the benefit-to-risk ratio that physicians and their patients need to understand."

The JUPITER trial was the first placebo-controlled clinical trial to formally document an increased risk of diabetes in patients treated with statin therapy. For those treated with rosuvastatin (Crestor, AstraZeneca), 270 new cases of physician-diagnosed diabetes developed, compared with 216 cases of incident diabetes in the placebo-treated patients, a statistically significant 27% relative increase in risk.

In addition, as previously reported by heartwire , multiple meta-analyses confirmed the increased risk with statins in JUPITER, with an analysis from the Women's Health Initiative (WHI) showing a 48% increased risk of diabetes among women, while an analysis of PROVE-ITA to ZTNTIDEAL, and SEARCH showed that high-dose statin therapy increased the risk of diabetes by 12%. In addition, a similar meta-analysis found that statin therapy increased the risk of diabetes by 9%.

As a result of these studies, the US Food and Drug Administration (FDA) changed the label for the entire drug class to warn physicians that statins can raise blood sugar and glycosylated HbA1c levels.

To heartwire , Dr Kausik Ray (St George's University of London, UK), who has been involved in multiple analyses of clinical trials looking at the diabetes risk from statins, said the JUPITER results are consistent with previous studies showing a benefit of statin therapy in high-risk patients but offer insight into the primary-prevention population. The concern has been that the risk of diabetes would offset the benefits in patients at lower risk for cardiovascular disease.

This is a really important clinical issue to get right.

"What this suggests is that in healthy people who do not yet have diabetes and who are at risk for cardiovascular events for whatever reason, the number of cardiovascular events prevented is going to be greater than the risk of diabetes."

One caveat, however, is that the JUPITER study was stopped early, just before the two-year mark. When this is done, said Ray, who was not involved in the JUPITER analysis, the cardiovascular benefits tend to be overestimated while the risks of diabetes might be underestimated. "What we need now is a little bit longer follow-up," he added. "What these data show is that in the short term, the benefits clearly favor statin therapy."

Stratifying Patients Based on Their Baseline Risk for Diabetes

The JUPITER study included 17 603 men and women without previous cardiovascular disease or diabetes and randomly assigned them to treatment with rosuvastatin 20 mg or placebo and followed them for an average of 1.9 years. The primary results were reported previously, but briefly, investigators showed that treatment reduced the primary end point, a composite of nonfatal MI, nonfatal stroke, unstable angina, revascularization, and death from cardiovascular causes, by 44%. In the current analysis, Ridker and colleagues stratified patients into two groups: patients with one or more major risk factors for diabetes (n=11 508) and those without any risk factors (n=6095).

What these data show is that in the short-term, the benefits clearly favor statin therapy.

In patients without any risk factors for diabetes, statin therapy was associated with a significant 52% reduction in the risk of the primary end point, a significant 53% reduction in the risk of venous thromboembolism (VTE), and a significant 22% reduction in total mortality. Similarly, treatment with rosuvastatin among patients with one or more diabetic risk factors significantly reduced the primary end point 39%. Although there was a 36% and 17% reduction in the risk of VTE and total mortality, the reductions were not statistically significant.

In terms of the diabetes risk, statin therapy significant increased the risk 28% compared with placebo among individuals with risk factors for diabetes at baseline (hazard ratio 1.28; 95% CI 1.07–1.54). In contrast, there was no increased risk of diabetes among those without baseline risk factors.

"What we found was clinically important," said Ridker. "Among those with one or more major risk factors for diabetes, there were 134 fewer heart attacks, strokes, and other major cardiovascular events in those who got the statin, but this came with the hazard of 54 new cases of diabetes being diagnosed. This group is already at high risk for getting diabetes, a group already considered candidates for statin therapy."

For Ridker, the bottom line is a fairly simple, straightforward clinical message: "Even in a lower-risk primary-prevention population, the cardiovascular benefits of statin therapy outweigh the diabetes hazard, even among those with highest risks for diabetes."

To heartwire , Ray said that one of the difficulties for physicians in practice is applying the results of clinical trials to individual patients. In addition, use of different drugs and different dosages alter the risk/benefit ratio. For example, rosuvastatin 20 mg reduces LDL cholesterol by approximately 50%, which translates into a larger reduction in cardiovascular events. Weaker statins would not reduce LDL cholesterol to the same extent, but the risk of diabetes would be unaltered, and this would change the risk/benefit equation, say Ray.

"JUPITER is the largest primary-prevention [trial] ever done," he told heartwire . "It has 40% women and has representation from diverse ethnic groups, so it's truly a global study. It provides considerable reassurance about net benefit. But many of the people whom we treat with statins won't go on to have an event. The reason we use them is that, by and large, the treatments are actually pretty safe and pretty good. But if we could further improve risk prediction accuracy, then we could really tell patients that their risks are very high and that they would benefit from the drugs. In somebody else, if we could say to them, 'You're not at very high risk over the long term,' and we could avoid unwanted treatment. This is the unmet need that hasn't been addressed by this paper."

Data From General Taiwanese Population

In a paper published online August 8, 2012 in the Journal of the American College of Cardiology, Dr Kang-Ling Wang (National Yang-Ming University, Taipei, Taiwan) and colleagues also tackled the risk of diabetes with statin therapy, this time in the general population [2].

Using data from a national health insurance database, they found that the risk of diabetes was 2.4% in men >45 years of age and women >55 years of age taking statins compared with 2.1% in statin nonusers (p<0.001). During the more than seven years of follow-up, they report that statin therapy increased the risk of diabetes 15% (hazard ratio 1.15; 95% CI 1.08–1.22). Regarding benefit, the researchers report that statin users had a significant 18% reduction in the risk of MI, a significant 9% reduction in the risk of major adverse cardiac events, and a 39% lower risk of in-hospital mortality.

In their analysis of risk and benefit, Wang and colleagues concluded that use of statin therapy is most beneficial in high-risk and secondary-prevention patients. They add that outcomes of patients who developed diabetes after statin use were generally favorable, but continuous surveillance of dysglycemia should be included in management of patients treated with the lipid-lowering agents.

Value of Statins in Primary Prevention

In an editorial accompanying the JUPITER study [3], Drs Gerald Watts and Esther Ooi (University of Western Australia, Perth) said the analysis "reaffirms the new value of statins in the primary prevention of cardiovascular disease but intuitively cautions that the diabetogenic effect of these drugs is highest in individuals with risk factors for diabetes, including C-reactive protein [CRP]." In the absence of diabetes risk factors, "the risk of statins is negligible," they add. What remains to be seen is whether or not the conclusions about the risk of incident diabetes apply to subjects without elevated CRP levels.

In the absence of diabetes risk factors, the risk of statins is negligible.

Importantly, the editorialists question whether or not the effect of statins on new-onset diabetes is causal, ultimately suggesting that it might be. "The consistency of the clinical-trial findings, the temporality of the association, the dose-dependency of the effect, and the plausible mechanisms all provide a strong argument for causality," they write. "Despite persuasive data supporting biological plausibility, further work is needed to elucidate the mechanisms of action at hepatic, skeletal-muscle, adipocyte, beta-cell, and mitochondrial levels that could impair insulin action and glucose homeostasis."

Regarding which patients might benefit most from statin therapy, Watts and Ooi suggest specifically targeting patients without risk factors for diabetes, as this approach could provide greater gains without increasing the risk of diabetes. To heartwire , Ridker said that what matters most for optimal patient care is the balance of risk and benefit. "We hope these new data will better inform discussions between physicians and patients who are considering the use of statin drugs as a possible addition to diet, exercise, and smoking cessation," he said.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.