Lipitor, a statin, is used to treat high cholesterol
It is well known that the cholesterol-lowering drugs called statins can reduce the risk of heart attack among people who already have heart disease. But whether the medications can prevent a heart attack from occurring in the first place is still a hotly contested question among health experts.
Two new studies published on Monday in the Archives of Internal Medicine reignite the simmering debate.
One study revisits the merits of the controversial Jupiter trial (or Justification for the Use of Statins in Primary Prevention), which was published in 2008 in the New England Journal of Medicine. That trial concluded that the statin drug Crestor (rosuvastatin) lowered the combined risks of heart attack, stroke, other heart events or heart-related death by 47% in healthy patients with no history of heart problems or high cholesterol but high levels of C-reactive protein (CRP), a marker for inflammation. The findings prompted the Food and Drug Administration in February to expand the eligible patient population for Crestor by millions.
Critics of the Jupiter trial have contended that the benefits of the cholesterol drug may have been exaggerated because the trial was stopped early — after two years, instead of the planned five. Had the trial been allowed to continue, critics say, the differences in benefit between the treatment and placebo groups may have disappeared. That is the argument raised again by the new study in the Archives, by an international group of scientists led by Dr. Michel de Lorgeril at the University Joseph Fourier and the National Center of Scientific Research in Grenoble, France.
Jupiter was stopped prematurely when an independent monitoring board gleaned an overwhelming treatment benefit in the statin group. Although the early termination of randomized and blinded control studies is common — to ensure the safety of patients, study leaders frequently monitor the accruing data and stop the trial when one group shows a predetermined amount of benefit over the other — in Jupiter's case, de Lorgeril's group argues, the study never made clear what the predetermined benefit was.
What the data did show, however, is that when certain hard clinical endpoints — such as heart-related death — were considered, the difference between the two groups was not significant enough to warrant stopping the trial. Among the entire study population of more than 17,000, there was a total of only 240 heart attacks and strokes, a relatively small number; of those, the number of fatal heart attacks and strokes in the statin and placebo groups was equal, at 12 each, suggesting no benefit of the drug.
Still, when other endpoints were taken into consideration, the benefit of Crestor blossomed. Patients taking Crestor had significantly lower rates of nonfatal events: the treatment group showed 65% fewer nonfatal heart attacks, compared with people taking placebo, and a 48% lower rate of nonfatal stroke. Although a reduction in nonfatal cardiovascular events is important, and certainly desirable, heart experts argue that the true measure of a therapy is its effectiveness in extending the lives of people with disease.
Jupiter's principal author, Dr. Paul Ridker, agrees, but explains that because his patient population was healthier and necessarily had a lower rate of heart events, compared with patients who are typically recruited for statin studies, Jupiter would have had to continue for an unjustifiably long time in order to document a difference in heart-related death rates. "When a class of drugs is well recognized to reduce nonfatal heart attacks and strokes, then unless something truly unusual is happening, of course it will reduce fatal events, if you ran the trials long enough," says Ridker, a cardiologist at Brigham and Women's Hospital in Boston.
De Lorgeril and his co-authors suggest it would have been better to let the trial continue, and find out.
The second paper appearing in the Archives is a review of data from 11 previous randomized controlled trials of statins — including Jupiter — in healthy populations at high risk for heart disease. The study, led by Dr. Kausik Ray, who was then at St. George's, University of London, found that among more than 65,000 healthy people with no history of heart problems but with traditional risk factors for heart disease, treatment with the cholesterol-lowering medications did not significantly lower the risk of death from any cause over an average of four years.
Previous meta-analyses had shown that statins do in fact extend life, but those studies included patients with pre-existing cardiovascular disease — patients whom statins are known to benefit — potentially skewing the results, explains Ray. His current analysis looked only at healthy, high-risk individuals. And given the lack of benefit seen in these patients, Ray cautions against using statins — which carry serious side effects — for the primary prevention of heart disease in lower-risk populations.
Ridker believes such reluctance on the part of scientists to expand statin therapy to a larger patient population is unjustified. He notes that while Jupiter does call for the inclusion of new patients who are not treated under current statin guidelines, it does not support blanket treatment in "lower risk" patients: Jupiter's patients did not have the traditional known risk factors for heart disease (high blood pressure and high cholesterol, for instance), but they did have high levels of CRP, which indicate inflammation in the heart vessels. The trial was designed to test the theory that such inflammation could be a novel — and treatable — risk factor for heart attack and stroke that current guidelines do not take into account. He says Jupiter's results support the theory.
"I agree with the [Archives study] authors' bottom line, which is what we have always said over and over: the first things to do to prevent heart disease is to eat a proper diet, exercise regularly and stop smoking," says Ridker. "But what Jupiter clearly shows is that even among people who are thin, who exercise, have low cholesterol and don't smoke, the risk is high if you have elevated levels of CRP. And being on a statin can lower your risk of having a heart event by half."
So what does this mean for statin use in healthy individuals? Current guidelines suggest that people whose risk of having a heart event is at least 20% over the next 10 years — calculated by considering traditional risk factors such as a person's age, high cholesterol, elevated blood pressure and diabetes, but not high CRP — may benefit from taking the drug.
Ray agrees that these high-risk patients should continue using statins but says the medication should not be extended beyond this group. "If we crop-dusted the entire population and extended the number of people taking statins, we would get even less mortality benefit," he says. "We are not particularly brilliant in the primary prevention setting in working out who best benefits [from statin use]. We are still not perfect in the way we identify those individuals."
Ridker agrees that the system isn't perfect but counters that we now have at least one additional risk factor that can be addressed by a statin: inflammation. He thinks that people with high CRP should be considered for statin therapy, even if they don't show the traditional heart disease risk factors. "We didn't simply demonstrate that statins are highly effective in primary prevention," he says. "We also demonstrated that if you take a population of people who never qualified for a statin before, and who by current guidelines are considered to be at low risk of heart disease, and screen them for CRP, their risk is actually high."
"One thing to emphasize here is that what we are witnessing now is a statistical tug-of-war," says Dr. Clyde Yancy, president of the American Heart Association. "The broader message continues to be straightforward. We understand that we have new risk markers, and we should acknowledge that. And we understand that for some persons, the current indications for initiating statin therapy are too conservative and don't allow us to catch everyone. But in the same breath, we also recognize that giving someone a potentially lifelong therapeutic intervention when they are asymptomatic needs to be respected at the highest order."
The tension and confusion over making such treatment decisions runs deep. In a special article accompanying the two studies in the Archives, Dr. Sanjay Kaul, a cardiologist at Cedars-Sinai Medical Center, and colleagues conclude that Jupiter's findings are likely flawed and not sufficient to recommend using CRP to guide heart disease treatment decisions.
Time, and more data, may be the only way to resolve the debate. Already, millions of patients with elevated CRP levels are taking statins, so researchers will inevitably accumulate additional lifetime data on whether the drugs can reduce healthy people's risk of dying from heart disease.