Generic Lipitor: New Era for Statins? Statins for Everyone Over 35 Yrs Old?
By Chris Kaiser, Cardiology Editor, MedPage Today
Published: November 30, 2011|
After a long and successful run as a branded statin, Lipitor (atorvastatin) became available as a generic on Wednesday.
In an interview earlier this month at the American Heart Association meeting, Donald Lloyd-Jones, MD, of Northwestern University in Chicago, told MedPage Today that the availability of generic atorvastatin could have a dramatic impact on cardiovascular care in the U.S., and many others have predicted that the generic could help usher in a new era of more widespread prescribing of statins.
Benefits and Side Effects
Besides statins' ability to lower LDL cholesterol, they have been praised for other beneficial effects including, for example, conferring a protective benefit to the brain in head trauma and to the heart in those with atrial fibrillation.
Just as important, however, is what statins won't do, since they have been associated - both correctly and incorrectly - with various unwanted side effects. But a recent study seems to have put to rest the idea that statins cause cancer.
Meanwhile, there have been reports about increased muscle injury with high doses of simvastatin (generic and branded [Zocor]), particularly when taken with certain other medications. The FDA in June outlined label changes and dose limitations for the drug.
In an email to MedPage Today and ABC News, William Golden, MD, director of general internal medicine at the University of Arkansas for Medical Sciences in Little Rock, said that the FDA warnings with simvastatin, "previously the most potent generic statin, make the arrival of generic Lipitor more important now than a couple of years ago."
He added that many of his patients are on "every-other-day Crestor [rosuvastatin] or Lipitor to manage musculoskeletal side effects of high-dose generic statins" and that generic Lipitor will become his choice for patients who fail pravastatin (Pravachol) management.
Expanding the Market
Earlier this month, the American Academy of Pediatrics issued new guidelines calling for cholesterol screening in all children between the ages of 9 and 11 and again at 17 to 21. Previous guidelines called for screening only those with risk factors.
The study that set the stage for statins to be widely prescribed beyond those with high cholesterol was JUPITER. This trial found that individuals with low LDL cholesterol but high-sensitivity C-reactive protein (hsCRP) benefited from statins with reductions in myocardial infarction and stroke.
JUPITER has been criticized, but overall the results appear to have withstood the test of time. The trial was referenced in a Perspective published online this month in Nature Reviews: Cardiology that outlined reasons why it would be prudent to prescribe statins to asymptomatic adults between the ages of 35 to 50.
"This notion of vitamin-like use of LDL-cholesterol-lowering therapy by young, asymptomatic individuals is a substantial departure from current practice. However, were application of this approach to the coronary heart disease epidemic to be successful, it would be a public health breakthrough of immense consequence -- a 'game changer,'" wrote Valentin Fuster, MD, from Mount Sinai School of Medicine, and his three Perspective co-authors.
Earlier this year, the FDA approved rosuvastatin for primary prevention of cardiovascular disease in people with high levels of hsCRP, a decision primarily based on the JUPITER trial results. It is the first statin to receive approval for such an indication.
With the availability of generic Lipitor, the stage seems to be set more than ever for the widespread use of statins. But strategies and guidelines have to be hammered out before physicians begin prescribing statins to many millions of asymptomatic individuals.
Physician Response Mixed
In the meantime, what is happening in clinical practice? Are clinicians gearing up for the wholesale switch from branded to generic atorvastatin? Some are, some aren't.
Marjorie A. Bowman, MD, MPA, from the department of family medicine and community health at the University of Pennsylvania in Philadelphia, indicated that her patients probably are not aware of generic Lipitor. "However, if it lowers the price, I am all for it, and my patients would be as well," she said in an email to MedPage Today and ABC News.
The scenario is a bit different for Roger Blumenthal, MD, director of the Johns Hopkins Ciccarone Preventive Cardiology Center. "Several patients have emailed me that they want to be switched from brand-name Crestor to generic atorvastatin," he said in an email. "The recent publication of the SATURN trial would suggest that generic atorvastatin did nearly as well as rosuvastatin (Crestor) did in achieving modest reversal of atherosclerotic plaque."
Typically, after a drug goes off patent, one generic brand is available for a few months at about 25% of the cost of the branded version. Then a few more generic brands come on the market and the price drops even lower.
"The real change will be six months from now when more than one company can sell the generic, and it becomes as inexpensive as simvastatin, pravastatin and lovastatin [Mevacor], which all are $4 per month at Wal-Mart and other retailers," said Christopher P. Cannon, MD, from Brigham and Women's Hospital in Boston, in an email to MedPage Today and ABC News.
"The issue is that it takes cost totally out of the equation in the discussion of whether to use a statin, and whether to use high-dose statins for patients who are hospitalized," he said. "The biggest change may be from primary care physicians who hesitate to use atorvastatin 80 mg. Most would use 40 mg (for general fear of using high doses of statins), but if they adopt atorvastatin more, it will help drive down LDL levels."
Brand-Name Company Holding On
A few reports have indicated that Pfizer, the maker of Lipitor, is not quietly yielding to the eventual demise of its multi-billion dollar cash cow. The company has continued aggressive marketing of the drug and has created partnerships with insurers to keep its drug in the limelight.
In addition, the company has offered customers steep discounts to continue to take Lipitor.
How all this plays out remains to be seen, but it could be that Pfizer has tapped into consumer fears of switching from a known entity to something less known.
"Patients are awaiting the generic version because the brand-name drug is expensive. They ask if the drug will be the same and are concerned about side effects," said Nieca Goldberg, MD, medical director of NYU Women's Heart Program in New York City.
Those sentiments were echoed by Randal J. Thomas, MD, MS, from the Mayo Clinic's Cardiovascular Health Clinic in Rochester, Minn. "One area of possible concern with generic medications is the possibility that the generic atorvastatin options will be different in potency compared to Lipitor brand atorvastatin," he wrote in an email to MedPage Today and ABC News. "This difference is likely to be minimal, but it will be important for patients who switch from brand-name Lipitor to generic atorvastatin to follow up with blood tests and a check-up with the healthcare provider to make sure that the atorvastatin dose is the right one for them."
Philip Ades, MD, director of cardiac rehabilitation at the University of Vermont Medical School in Burlington, took issue with Pfizer's approach.
"That Pfizer is trying to squeeze every nickel out of Lipitor is not surprising, but it is at the heart of why medical care costs so much in the U.S. On the one hand, everyone bemoans the high cost of healthcare; on the other hand, the business model of 'making money off of our patients' and maximizing profit is not compatible with controlling costs and focusing on outcomes.
"You cannot have both high-profit and low-cost outcomes-driven care. If healthcare is deemed a right and not a privilege, industry profits need to be regulated," he said in an email to MedPage Today and ABC News.
Ultimately, however, the answer to whether patients should switch statins is fairly simple, according to Cannon. "Whatever is lowest cost and gets the LDL down is okay."
Opinion: Can we dramatically reduce the incidence of coronary heart disease?
Nature Cardio Reviews Dec 2011
Michael Domanski, Donald Lloyd-Jones, Valentin Fuster & Scott Grundy
Mount Sinai School of Medicine, One Gustave L. Levy Place, New York, NY 10029, USA (M. Domanski, V. Fuster). Northwestern University Feinberg School of Medicine, 251 East Huron Street, Galter Suite 3-150, Chicago, IL 60611, USA (D. Lloyd-Jones). University of Texas Southwestern Medical School at Dallas, 5323 Harry Hines Boulevard, Dallas, TX 75390, USA (S. Grundy).
Correspondence to: V. Fuster email@example.com
"Compelling data suggest the possibility that dramatic reductions in the incidence of CHD and stroke, perhaps to the point of rendering them of limited relevance to public health, might be made possible by starting statin treatment across the population beginning in young to middle adulthood. Such a vitamin-like approach to CHD prevention can be tested in a randomized clinical trial of a size, duration, and cost comparable with those of current large, simple trials."
Thrombotic events caused by atherosclerosis are by far the most common cause of death in the world. Advances in treatment have prolonged the life of patients with atherosclerotic disease, but treating individuals with clinically manifest disease will not result in a cure, and will not prevent the majority of such events. This paper explores the possibility that early implementation of a drug-based approach to the prevention of atherosclerosis and atherothrombotic events could dramatically reduce the footprint of this disease. Specifically, we consider the potential impact on atherothrombotic events of a population-based, pharmacological approach to lowering serum LDL-cholesterol levels in asymptomatic adults aged 35-50 years. We also consider the feasibility and basic design considerations for a clinical trial to assess the validity of this public health and primary prevention approach.
In the lifetime of a child born today, atherothrombotic events will result in the deaths of more people than all of the wars of history.1, 2 Most of these events, particularly in the USA, are related to coronary atherosclerosis and atherothrombotic stroke. Indeed, atherothrombotic events are the most common cause of death in the world.1 Moreover, such events frequently result in disability and necessitate expenditure of massive fiscal resources-more than US$100 billion annually in the USA alone.3
Advances in treatment have increased the longevity of patients with coronary heart disease (CHD; defined for the purposes of this article as symptomatic coronary artery disease) and improved the quality of their lives. However, the prognostic implications of atherosclerotic disease mean that treatment of established CHD will never be able to change its status as the leading cause of death worldwide. The alternative to treating established CHD is to eliminate it by preventing or arresting the development of atherosclerosis. Is prevention a realistic goal now, or must we await grand new strides in our fundamental understanding of biological processes and how to manipulate them? The confluence of advances in our understanding of the mechanisms underlying CHD, data from epidemiological studies and clinical investigations, along with the availability of safe medication that dramatically lowers serum LDL-cholesterol levels, raises the intriguing possibility that prevention of CHD on a population (or even global) scale might be feasible. These efforts could include simple, currently available strategies; specifically, the early institution of lipid-lowering drug therapy even in individuals with no evidence of vascular disease.
This notion of vitamin-like use of LDL-cholesterol-lowering therapy by young, asymptomatic individuals is a substantial departure from current practice. However, were application of this approach to the CHD epidemic to be successful, it would be a public-health breakthrough of immense consequence-a 'game changer'. In this paper we consider the available data supporting the hypothesis that early initiation of such therapy will dramatically reduce the incidence of CHD. We also discuss the feasibility of designing a randomized, controlled trial to assess the efficacy of this intervention, such that positive results would compellingly support a guideline recommendation for instituting pharmacological LDL-cholesterol lowering in asymptomatic individuals during early middle age (35-50 years).
Serum LDL as a driver of CHD
The atherosclerotic process is initiated by LDL entry into the vessel wall, which is, in turn, driven by the serum LDL-cholesterol concentration.4 As might be expected, epidemiological data demonstrate a strong, graded association between serum LDL-cholesterol concentrations and CHD event rates.5, 6, 7, 8, 9 Pathology and imaging studies also show a strong association between the serum LDL-cholesterol level and the presence and extent of coronary atherosclerosis, which confirms the importance of this mechanism4, 5 and provides insight into the physiology underlying these findings.4
Data from studies of hunter-gatherer populations, who have serum LDL-cholesterol concentrations of 1.30-1.81 mmol/l,3 provide additional evidence that lifelong low serum levels of LDL cholesterol have a cardioprotective effect.4, 10, 11, 12, 13, 14 These populations generally exhibit little CHD as long as their traditional lifestyle is maintained.10 However, as individuals from these cultures adopt a modern, western diet and lifestyle, their serum LDL-cholesterol levels increase and CHD becomes commonplace.13
Although indicative, the strong association between serum LDL-cholesterol concentration and the incidence of CHD events in a population does not by itself prove that therapeutic lowering of LDL-cholesterol levels will reduce the CHD event rate. For this reason, prospective, randomized clinical trials have been performed to assess the therapeutic role of lowering serum LDL-cholesterol levels in the primary prevention of CHD.15, 16, 17, 18, 19 All these studies show a strong, graded association between the serum LDL-cholesterol level achieved by patients taking the study drug and the CHD event rate. JUPITER20, 21 is notable in that patients had modest serum LDL-cholesterol levels at entry into the study (median 2.80 mmol/l) and achieved remarkably low serum LDL-cholesterol levels by treatment with rosuvastatin (median 1.42 mmol/l at 12 months). The results showed a reduction in the primary composite end point of myocardial infarction (MI), stroke, unstable angina, arterial revascularization, or cardiovascular death in statin-treated patients. A subsequent post hoc analysis showed a dramatically lower rate of the primary end point in the subgroup of JUPITER participants who achieved serum LDL-cholesterol levels <1.30 mmol/l than in the overall group randomly assigned to treatment with rosuvastatin.21 Importantly, none of the studies conducted thus far have identified a serum LDL-cholesterol level below which no further reduction in the CHD event rate occurs-the lower the better, at least for the serum LDL-cholesterol levels that have been achieved in clinical trials.4, 5
The case for early intervention
Clinical trials to date have dealt largely with populations aged over 50 years. Although the therapeutic value associated with lowering of serum LDL-cholesterol levels has been unequivocally demonstrated in these populations, substantial residual CHD risk is evident even in the active-treatment groups. This risk would probably have been mitigated by achieving even lower LDL-cholesterol levels with therapy. However, the residual risk is probably also influenced by the atherosclerotic burden already present in these patients at the start of these studies-a burden that would have been either reduced or absent altogether if statin treatment had been started earlier.
Data from the ARIC study22 provide strong evidence of the importance of maintaining a consistently low LDL-cholesterol level throughout life.23 In this analysis, Cohen et al. examined the effect of nonsense mutations in the PCSK9 gene, which encodes the serine protease, proprotein convertase subtilisin/kexin type 9;23 inactivation of this gene leads to reduced serum LDL-cholesterol concentrations. These mutations were found in 2.6% of middle-aged (45-64 years) black participants and were associated with 28% lower serum LDL-cholesterol concentrations on average. The mutations were also found in 3.2% of middle-aged white participants and were associated with a 15% lower serum LDL-cholesterol level on average. The black participants with these mutations had an 88% lower risk of CHD events, and the white individuals a 50% lower risk of CHD events over 15 years of follow-up, compared with individuals of the same ethnic groups who did not carry these mutations.23 These CHD event rates are far lower than would be expected, on the basis of data from primary prevention clinical trials, from the modest reductions that these individuals had in their serum LDL-cholesterol levels (Figure 1). This apparent discrepancy might result from the fact that the reduced LDL-cholesterol concentration is present throughout the lifespan of individuals who carry these mutations. Furthermore, low serum levels of LDL cholesterol, resulting from genetic variants or lifestyle, in childhood or young adulthood, have been associated with a decreased burden of subclinical atherosclerosis by middle age.24
A meta-analysis of randomized trials showed that the percentage CHD risk reduction associated with a decrease of 0.26 mmol/l in serum LDL-cholesterol levels declines for each decade of life after the fifth (the researchers did not have data for the first to fourth decades).25 The obverse, therefore, is also true-the benefits of LDL-cholesterol-lowering therapy are greater the earlier that such therapy is initiated (at least for individuals aged over 50 years). The question that the results of this study fails to answer, however, is how lifetime CHD risk is affected by lipid-lowering therapy at ages below 50 years. The possibility exists that even greater reductions in CHD risk could be achieved, perhaps even approaching the very low absolute lifetime risk of ~5% demonstrated for individuals who have the best possible cardiovascular risk profile at age 50 years,24 by starting therapy in young adulthood.
Synthesis of the available data from epidemiological studies and clinical trials (Figure 1) leads to the hypothesis that early, profound lowering of LDL-cholesterol levels might dramatically reduce the CHD event rate across the remainder of the individual's lifespan. How low should target levels be, and how early should therapy be initiated? Human neonates and hunter-gatherer populations have serum LDL-cholesterol levels of 1.30-1.81 mmol/l, so this range of serum LDL-cholesterol concentrations seems to be a reasonable goal. However, despite the strong suggestion of benefit accrued by the foregoing evidence, as well as computer simulations indicating its efficacy26 and cost-effectiveness,27 the willingness to expose a large, asymptomatic population of young adults to long-term pharmacological lipid-lowering therapy will require a clear demonstration of benefit in a suitably powered, randomized clinical trial.
A rational study design
To be practicable, a randomized clinical trial that aims to validate a strategy of treating healthy asymptomatic young or middle-aged adults must have a similar duration to that of the therapeutic trials of statins described above (7-10 years) and have a sample size that is not substantially greater than those of these large, simple trials (~20,000 patients).15, 16, 17, 18, 19, 20, 21 However, the study must also have sufficient power to generate definitive evidence for a guideline recommendation for early pharmacological lowering of LDL-cholesterol levels in asymptomatic, young individuals. These considerations provide boundary conditions that constrain the effect size that is reasonable to resolve. We posit that the effect size needed to justify giving statin therapy to a large number of healthy, asymptomatic young to middle-aged individuals needs to be larger than that required in most randomized clinical trials. Specifically, the effect size, we believe, must be around 50% or more to be clinically relevant and demonstrate adequate public health benefits. This condition would result in a sample size well within the range of current large clinical trials which, as noted above, is important to ensure feasibility. Taking into account these considerations, we address the central questions that relate to trial design, below (Box 1).
What age range is appropriate?
The JUPITER data, obtained in patients with a mean age of 60 years, suggest that substantial residual CHD risk remains after statin treatment in the context of a lifetime of exposure to high serum levels of LDL-cholesterol. What is not yet known is how the age at which hunter-gatherer levels of serum LDL-cholesterol are achieved affects CHD risk. For instance, what would be the lifetime CHD risk for healthy, asymptomatic individuals who achieved LDL-cholesterol levels of 1.30-1.81 mmol/l pharmacologically between the ages of 35 years and 45 years? If a 50% reduction in CHD risk were indeed demonstrated over the relatively short duration of the trial, the result would be a sea change in the public health implications of CHD. Owing to the very large potential benefit combined with the practical problem of avoiding a trial of greater than 10 years' duration, we believe that this is the appropriate age range for participants at entry to the study. Specifically, we propose that men aged 35-45 years and women aged 35-59 years who do not have child-bearing potential (since statins can have teratogenic effects)28 should be included.
Who should be excluded?
Women aged 60 years or older were studied in JUPITER, so this age group can be excluded. In addition, the exclusion of women of child-bearing potential means that female participants in the trial will mostly be 50-59 years old. Although at first glance this difference would seem to create a disparity between male and female participants, the fact that women tend to develop atherosclerotic disease ~10 years later than men might mitigate its effect on the results.
Individuals with an LDL-cholesterol level high enough to meet current guideline criteria for immediate treatment (that is, ≥4.92 mmol/l) would, of course, be excluded from the trial. The more difficult question is whether a lower limit of LDL-cholesterol levels exists, below which an individual should be excluded because further reduction is unnecessary. However, given the very low CHD risk of individuals with an LDL-cholesterol level <1.81 mmol/l, exclusion of these individuals seems reasonable. In addition, exclusion of people with a very low short-term and lifetime risk of major cardiovascular events (such as those with optimal cholesterol levels and blood pressure, no diabetes, and no smoking history) also seems prudent, since they are extremely unlikely to benefit from lipid-lowering therapy in the short or long term.
What sample size would be needed?
In a hypothetical study involving men aged 35-45 years (mean age 40 years) and women aged 35-59 years (assuming a mean age of 50 years, resulting from the need to exclude women of childbearing potential), currently available epidemiological data predict that the event rate (cardiovascular death, MI, or stroke) would be approximately 4% for a trial of 10 years duration with a mean follow-up of 8.5 years.29, 30 Assuming, for illustrative purposes, that the study population comprised 66% male and 33% female participants, and had similar rates of cross-over and loss to follow-up to other statin trials, the total sample size required would be approximately 20,000 individuals to provide 85% power to detect a 50% reduction in such events during the trial. This sample size is similar to that of JUPITER and other large clinical trials of statin therapy.15, 16, 17, 18, 19, 20, 21
The effects of a positive outcome
Two lines of argument underlie why we believe that positive results for such a trial would be sufficiently compelling to result in a guideline recommendation to treat asymptomatic adults. Firstly, a reduction of 50% in the rate of death, MI, and stroke, with or without revascularization (the composite primary end points that we propose) would itself justify a guideline recommendation for treatment. Approximately 40.5 million people in the USA are currently aged within the ranges suggested as inclusion criteria. An estimated cardiovascular event rate of 4% over an average of 8.5 years translates into approximately 1.38 million deaths, MIs, or thrombotic strokes. If half these events were prevented, more than 81,000 deaths, MIs, or strokes would be avoided every year in the USA alone. The real benefit of lipid-lowering treatment would be immense, if this strategy proved successful and were applied worldwide.
Secondly, the long-term benefits of early treatment might substantially exceed those seen during the course of the trial. For example, data from the Cholesterol Treatment Trialists' analysis of individual-level data from numerous statin trials31, 32 demonstrate that the long-term benefits of statin therapy (namely, relative risk reductions and a progressive increase over time in the absolute benefit, indicated by reduced rates of CHD and major cardiovascular events) are at least equal to, if not greater than the short-term benefits, even in those with no prior cardiovascular disease. Furthermore, data from long-term follow-up of participants in WOSCOPS29 provide compelling evidence for the persistence of these beneficial effects, at least in middle-aged men (aged 45-64 years at enrollment). During the 4.8 years of follow-up in this study, statin therapy was associated with a 31% relative risk reduction in death from CHD or nonfatal MI (the absolute event rates were 7.9% with placebo versus 5.5% with pravastatin 40 mg daily).27 At the end of the trial, the study drug was no longer provided to participants, and by 5 years after completion of this trial, 35.2% of the placebo group and 38.7% of those who had been assigned to pravastatin were being treated with a statin.29 Nonetheless, despite these similar post-trial statin treatment rates, by 15 years after randomization the event rates for death from CHD or nonfatal MI were still significantly different-15.5% in the group originally randomized to placebo and 11.8% in those originally randomized to pravastatin (P <0.001).29 These data strongly suggest that early initiation of statin therapy might have important effects in preventing and delaying clinical events for a long period of time, probably through arresting or stabilizing the development or progression of atherosclerotic plaque. Event-free survival curves continue to separate over time in trials of statin treatment versus placebo, which suggests that the effect size is likely to increase with continued follow-up, rather than decrease.31, 32 The public health impact of a 50% reduction in event rates (not to mention those of a greater than 50% reduction) over 10 years would, therefore, be immense in the long term.
A further consideration is the long-term safety of statins. Although some patients cannot tolerate these drugs, serious adverse events (specifically, rhabdomyolysis) are vanishingly rare in patients who do not have contraindications to statin treatment. Additionally, while the effects of very prolonged statin use (that is, over most of a lifetime) have not been studied, treatment with statins for 15 years or more has not been associated with notable late complications in the groups of individuals who we suggest should be included in the study.
Is the cost justifiable?
We suggest that the cost of such a trial, performed in the usual fashion with a series of clinical sites, a data coordinating center, and a clinical coordinating center, would be comparable with that of other large statin trials that had a similar sample size and did not require frequent or extensive patient contact. Given that fixed-dose statin regimens were used in existing studies, and that the expected effect of these doses on LDL-cholesterol levels is known, follow-up blood tests could be limited to those needed for safety monitoring and clinical contact could be minimized to save costs.
We believe that this study would be amenable to execution via new technologies that make use of web-based recruitment of participants and follow-up, with the trial interface being conducted largely via nonphysician personnel. For instance, initial screening for eligibility could be done online, and a nurse visit could then be scheduled for recruitment, obtaining informed consent, and blood drawing. Follow-up could include medication mailed to the patient with remote physician supervision of the process. The cost of the trial, using such a model, would be much less than that of prior studies based exclusively at clinical centers.
Atherothrombotic events are the number one cause of death worldwide. Compelling data suggest the possibility that dramatic reductions in the incidence of CHD and stroke, perhaps to the point of rendering them of limited relevance to public health, might be made possible by starting statin treatment across the population beginning in young to middle adulthood. Such a vitamin-like approach to CHD prevention can be tested in a randomized clinical trial of a size, duration, and cost comparable with those of current large, simple trials.