- Email: [email protected]
Is Rosuvastatin Really a More Efficient Therapeutic Option than Atorvastatin?
Blackwell Publishing IncMalden, USAVHEValue in Health1098-30152006 Blackwell Publishing200694276279Letter to the EditorLetter to the EditorLetter to the Editor
Volume 9 • Number 4 • 2006 VALUE IN HEALTH
Is Rosuvastatin Really a More Efficient Therapeutic Option than Atorvastatin?
To the Editor–With regard to an article recently published in this journal, we would like to state our disagreement with the conclusion drawn. The conclusion stated that rosuvastatin dominates atorvastatin because it is more effective and less costly [1]. We disagree with this conclusion because this cost-effective analysis was performed by using just surrogate end points without clinically meaningful outcomes and did not incorporate important safety data about rosuvastatin. It is very clear at present that when using statins the most important outcome is neither the reduction of lowdensity lipoprotein cholesterol (LDL-C) levels nor the increase of high-density lipoprotein cholesterol levels (both surrogate end points) but rather the decrease of hard end points, such as myocardial infarction, hospitalization for unstable angina, stroke, death form cardiovascular causes and sudden death. Therefore, it is essential to realize the limitation of surrogate end points in assessing both the efficacy and also the potential harmful effects of new medicines [2]. We do think that the efficacy of new drugs should be based on clear and reliable proof of clinically relevant outcomes (cardiovascular events avoided, deaths averted, lives saved, etc.) and the end result of performing a costeffectiveness analysis with surrogate end points is a partial result that may confuse decision-makers and their determinations. This is particularly important if we take into account that every single statin has demonstrated benefit in terms of these clinically relevant outcomes in large randomized clinical trials, with the exception of rosuvastatin. In addition, there is a large body of evidence suggesting that LDL is not the only mechanism explaining the benefits of statins. The so-called pleiotropic effects should also be taken into account. This is very well illustrated with the results of two studies: the PROVE-IT and REVERSAL [3,4], comparing atorvastatin 80 mg and pravastatin 40 mg in patients with coronary heart disease. In the PROVE-IT study, the benefit of atorvastatin 80 mg in terms of morbimortality was demonstrated in patients with acute coronary syndromes. In the REVERSAL study, the benefit of atorvastatin 80 mg against pravastatin 40 mg couldn’t be solely explained by LDL. It was hypothesized that a specific effect of atorvastatin 80 mg on C-reactive protein (CRP) levels might be responsible for the unexplained benefit. Based on this, in this cost-effectiveness analysis it had been very useful the inclusion of other surrogate end 10.1111/j.1524-4733.2006.00111.x
276
point, CRP levels, as it is an inflammatory marker that has a strong predictive of future cardiovascular events [5]. In addition, the reduction of CRP levels is independent of the decrease of LDL-C levels and it has been demonstrated that there is a weak correlation between the percentage reductions in LDL-C and CRP levels [6], so the only reduction of LDL-C does not predict very well the decrease of cardiovascular events and death from cardiovascular causes. Although all statins diminish the levels of both LDL-C and CRP, these are mostly independent of each other and the fact of decreasing LDL-C levels does not mean that the decline of CRP levels will be also significantly. Atorvastatin has demonstrated that it significantly reduces CRP levels and in a higher quantity than other statins do [4,7] while the degree of CRP levels reduction is unclear in rosuvastatin [8]. Therefore, a plausible conclusion is that atorvastatin produces a higher reduction in hard efficacy end points than rosuvastatin. This debate will be resolved when the JUPITER trial is completed. At the moment this trial is ongoing and will determine whether long-term use of rosuvastatin can reduce CRP levels and the rate of coronary events [9]. This confirms our thoughts that the results of this costeffectiveness analysis should be considered with a lot of precaution and it would be very dangerous to make decisions with them. On the other hand, in a recent published article it has been described that rosuvastatin may produce higher rates of adverse events report (AER) (rhabdomyolysis, proteinuria, nephropathy, and renal failure) than atorvastatin, simvastatin, and pravastatin in common clinical practice in the United States [10]. This analysis captures the real-life population exposure collected in the Food and Drug Administration AERs system and offers advantages over safety data coming from controlled premarketing clinical trials. We believe that it is necessary to incorporate these data into the decision analytic model, to complete safety data from clinical trials. Patients treated with rosuvastatin in who appear an AER probably need either a dose reduction or even a withdrawal of this statin, so the final efficacy in reducing LDL-C levels would be lower. In this case, atorvastatin could be a more efficient option than rosuvastatin. Although leading to mayor methodological caveats the CRP and AER issues are not the most important problems of this economic evaluation. The fundamental limitation is based in a much more obvious rationale: when morbimortality benefits have been demonstrated for a number
© 2006, International Society for Pharmacoeconomics and Outcomes Research (ISPOR) 1098-3015/06/276 276–279
277
Letter to the Editor of drugs within a particular group, such as statins, as per large randomized clinical trials, a surrogate end point simply cannot be used to claim a superior efficacy for the only drug (rosuvastatin) without any evidence of efficacy in terms of morbimortality. Cost-effectiveness analysis is a very useful tool for decision-makers to allocate properly available resources. However, if it is not correctly conducted or it is performed with inappropriate data, it is possible that at the end decision-makers may make incorrect decisions with the consequent being ineffective allocation of resources. Therefore, we believe that rosuvastatin should demonstrate efficacy in terms of morbimortality before trying to demonstrate cost-effectiveness. For these reasons, we do think that this costeffectiveness analysis should be considered a preliminary exploratory exercise, pending on the results of morbimortality and safety of rosuvastatin from large randomized clinical trials. Otherwise, conclusions are potentially misleading for decision-makers.—Javier Soto, MD, PhD, and Jaime Fernandez de Bobadilla, MD, Health Outcomes Research, Medical Unit, Pfizer, Madrid, Spain.
References 1 Benner JS, Smith TW, Klingman D, et al. Cost-effectiveness of rosuvastatin compared with other statins from a managed care perspective. Value Health 2005;8:618–28. 2 Lonn E. The use of surrogate endpoints in clinical trials: focus on clinical trials in cardiovascular diseases. Pharmacoepidemiol Drug Saf 2001;10:497–508.
3 Cannon CP, Braunwald E, McCabe CH, et al. Comparison of intensive and moderate lipid lowering with statins after acute coronary syndromes. N Engl J Med 2004;350:1495–504. 4 Nissen SE, Tuzcu EM, Schoenhagen P, et al. Effect of intensive compared with moderate lipid-lowering therapy on progression of coronary atherosclerosis: a randomised controlled trial. JAMA 2004;291:1071– 80. 5 Ridker PM, Rifal N, Rose L, et al. Comparison of Creactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N Engl J Med 2002;347:1557–65. 6 Nissen SE, Tuzcu EM, Schoenhagen P, et al. Statin therapy, LDL-cholesterol, C-reactive protein, and coronary artery disease. N Engl J Med 2005;352:29–38. 7 Ridker PM, Morrow DA, Rose LM, et al. Relative efficacy of atorvastatin 80 mg and pravastatin 40 mg in achieving the dual goals of low-density hipoprotein cholesterol <70 mg/dl and C-reactive protein <2 mg/ L: an analysis of the PROVE-IT TIMI—22 trial. J Am Coll Cardiol 2005;45:1644–8. 8 Doggrell SA. Relating statin therapy to C-reactive protein levels. Expert Opin Pharmacother 2005; 6:1597–600. 9 Riker PM. Rosuvastatin in the primary prevention of cardiovascular disease among patients with low levels of low-density lopiprotein cholesterol and elevated high-sensitivity C-reactive protein: rationale and design of the JUPITER trial. Circulation 2003; 108:2292–7. 10 Alsheikh-Ali AA, Ambrose MA, Kuvin JT, et al. The safety of rosuvastatin as used in common clinical practice. A postmarketing analysis. Circulation 2005; 111:3051–7.
Blackwell Publishing IncMalden, USAVHEValue in Health1098-30152006 Blackwell Publishing200694278279Letter to the EditorLetter to the EditorLetter to the Editor
The authors reply . . .
Formulary Decision-Making Should Rely on the Best Available Evidence
To the Editor—Evidence of effectiveness and costeffectiveness is inherently dynamic in nature. Costeffectiveness models based on the best data available at the time are an important contribution to evidence-based decision-making if they reveal the likely outcomes of various scenarios and quantify the degree of uncertainty around the apparently optimal approach. Cost-effectiveness models also illustrate the trade-offs inherent in each potential alternative available to the decision-maker and should be updated as new evidence becomes available.
10.1111/j.1524-4733.2006.00112.x
In 2000, The Academy of Managed Care Pharmacy issued its first Format for Formulary Submissions, which provided US payers with a new tool to proactively request clinical and economic data from health technology manufacturers. The Format, which has been revised substantially in recent years, is now used by health-care organizations covering some 150 million lives in the United States [1]. This guidance recommends that dossiers for new drug products be requested by health plans approximately 6 months before launch, and explicitly calls for the use of economic models “to inform decisions about the value or cost-effectiveness of pharmaceuticals, biologics, and vaccines.” Such models are to be based on
Volume 9 • Number 4 • 2006 VALUE IN HEALTH
Is Rosuvastatin Really a More Efficient Therapeutic Option than Atorvastatin?
To the Editor–With regard to an article recently published in this journal, we would like to state our disagreement with the conclusion drawn. The conclusion stated that rosuvastatin dominates atorvastatin because it is more effective and less costly [1]. We disagree with this conclusion because this cost-effective analysis was performed by using just surrogate end points without clinically meaningful outcomes and did not incorporate important safety data about rosuvastatin. It is very clear at present that when using statins the most important outcome is neither the reduction of lowdensity lipoprotein cholesterol (LDL-C) levels nor the increase of high-density lipoprotein cholesterol levels (both surrogate end points) but rather the decrease of hard end points, such as myocardial infarction, hospitalization for unstable angina, stroke, death form cardiovascular causes and sudden death. Therefore, it is essential to realize the limitation of surrogate end points in assessing both the efficacy and also the potential harmful effects of new medicines [2]. We do think that the efficacy of new drugs should be based on clear and reliable proof of clinically relevant outcomes (cardiovascular events avoided, deaths averted, lives saved, etc.) and the end result of performing a costeffectiveness analysis with surrogate end points is a partial result that may confuse decision-makers and their determinations. This is particularly important if we take into account that every single statin has demonstrated benefit in terms of these clinically relevant outcomes in large randomized clinical trials, with the exception of rosuvastatin. In addition, there is a large body of evidence suggesting that LDL is not the only mechanism explaining the benefits of statins. The so-called pleiotropic effects should also be taken into account. This is very well illustrated with the results of two studies: the PROVE-IT and REVERSAL [3,4], comparing atorvastatin 80 mg and pravastatin 40 mg in patients with coronary heart disease. In the PROVE-IT study, the benefit of atorvastatin 80 mg in terms of morbimortality was demonstrated in patients with acute coronary syndromes. In the REVERSAL study, the benefit of atorvastatin 80 mg against pravastatin 40 mg couldn’t be solely explained by LDL. It was hypothesized that a specific effect of atorvastatin 80 mg on C-reactive protein (CRP) levels might be responsible for the unexplained benefit. Based on this, in this cost-effectiveness analysis it had been very useful the inclusion of other surrogate end 10.1111/j.1524-4733.2006.00111.x
276
point, CRP levels, as it is an inflammatory marker that has a strong predictive of future cardiovascular events [5]. In addition, the reduction of CRP levels is independent of the decrease of LDL-C levels and it has been demonstrated that there is a weak correlation between the percentage reductions in LDL-C and CRP levels [6], so the only reduction of LDL-C does not predict very well the decrease of cardiovascular events and death from cardiovascular causes. Although all statins diminish the levels of both LDL-C and CRP, these are mostly independent of each other and the fact of decreasing LDL-C levels does not mean that the decline of CRP levels will be also significantly. Atorvastatin has demonstrated that it significantly reduces CRP levels and in a higher quantity than other statins do [4,7] while the degree of CRP levels reduction is unclear in rosuvastatin [8]. Therefore, a plausible conclusion is that atorvastatin produces a higher reduction in hard efficacy end points than rosuvastatin. This debate will be resolved when the JUPITER trial is completed. At the moment this trial is ongoing and will determine whether long-term use of rosuvastatin can reduce CRP levels and the rate of coronary events [9]. This confirms our thoughts that the results of this costeffectiveness analysis should be considered with a lot of precaution and it would be very dangerous to make decisions with them. On the other hand, in a recent published article it has been described that rosuvastatin may produce higher rates of adverse events report (AER) (rhabdomyolysis, proteinuria, nephropathy, and renal failure) than atorvastatin, simvastatin, and pravastatin in common clinical practice in the United States [10]. This analysis captures the real-life population exposure collected in the Food and Drug Administration AERs system and offers advantages over safety data coming from controlled premarketing clinical trials. We believe that it is necessary to incorporate these data into the decision analytic model, to complete safety data from clinical trials. Patients treated with rosuvastatin in who appear an AER probably need either a dose reduction or even a withdrawal of this statin, so the final efficacy in reducing LDL-C levels would be lower. In this case, atorvastatin could be a more efficient option than rosuvastatin. Although leading to mayor methodological caveats the CRP and AER issues are not the most important problems of this economic evaluation. The fundamental limitation is based in a much more obvious rationale: when morbimortality benefits have been demonstrated for a number
© 2006, International Society for Pharmacoeconomics and Outcomes Research (ISPOR) 1098-3015/06/276 276–279
277
Letter to the Editor of drugs within a particular group, such as statins, as per large randomized clinical trials, a surrogate end point simply cannot be used to claim a superior efficacy for the only drug (rosuvastatin) without any evidence of efficacy in terms of morbimortality. Cost-effectiveness analysis is a very useful tool for decision-makers to allocate properly available resources. However, if it is not correctly conducted or it is performed with inappropriate data, it is possible that at the end decision-makers may make incorrect decisions with the consequent being ineffective allocation of resources. Therefore, we believe that rosuvastatin should demonstrate efficacy in terms of morbimortality before trying to demonstrate cost-effectiveness. For these reasons, we do think that this costeffectiveness analysis should be considered a preliminary exploratory exercise, pending on the results of morbimortality and safety of rosuvastatin from large randomized clinical trials. Otherwise, conclusions are potentially misleading for decision-makers.—Javier Soto, MD, PhD, and Jaime Fernandez de Bobadilla, MD, Health Outcomes Research, Medical Unit, Pfizer, Madrid, Spain.
References 1 Benner JS, Smith TW, Klingman D, et al. Cost-effectiveness of rosuvastatin compared with other statins from a managed care perspective. Value Health 2005;8:618–28. 2 Lonn E. The use of surrogate endpoints in clinical trials: focus on clinical trials in cardiovascular diseases. Pharmacoepidemiol Drug Saf 2001;10:497–508.
3 Cannon CP, Braunwald E, McCabe CH, et al. Comparison of intensive and moderate lipid lowering with statins after acute coronary syndromes. N Engl J Med 2004;350:1495–504. 4 Nissen SE, Tuzcu EM, Schoenhagen P, et al. Effect of intensive compared with moderate lipid-lowering therapy on progression of coronary atherosclerosis: a randomised controlled trial. JAMA 2004;291:1071– 80. 5 Ridker PM, Rifal N, Rose L, et al. Comparison of Creactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N Engl J Med 2002;347:1557–65. 6 Nissen SE, Tuzcu EM, Schoenhagen P, et al. Statin therapy, LDL-cholesterol, C-reactive protein, and coronary artery disease. N Engl J Med 2005;352:29–38. 7 Ridker PM, Morrow DA, Rose LM, et al. Relative efficacy of atorvastatin 80 mg and pravastatin 40 mg in achieving the dual goals of low-density hipoprotein cholesterol <70 mg/dl and C-reactive protein <2 mg/ L: an analysis of the PROVE-IT TIMI—22 trial. J Am Coll Cardiol 2005;45:1644–8. 8 Doggrell SA. Relating statin therapy to C-reactive protein levels. Expert Opin Pharmacother 2005; 6:1597–600. 9 Riker PM. Rosuvastatin in the primary prevention of cardiovascular disease among patients with low levels of low-density lopiprotein cholesterol and elevated high-sensitivity C-reactive protein: rationale and design of the JUPITER trial. Circulation 2003; 108:2292–7. 10 Alsheikh-Ali AA, Ambrose MA, Kuvin JT, et al. The safety of rosuvastatin as used in common clinical practice. A postmarketing analysis. Circulation 2005; 111:3051–7.
Blackwell Publishing IncMalden, USAVHEValue in Health1098-30152006 Blackwell Publishing200694278279Letter to the EditorLetter to the EditorLetter to the Editor
The authors reply . . .
Formulary Decision-Making Should Rely on the Best Available Evidence
To the Editor—Evidence of effectiveness and costeffectiveness is inherently dynamic in nature. Costeffectiveness models based on the best data available at the time are an important contribution to evidence-based decision-making if they reveal the likely outcomes of various scenarios and quantify the degree of uncertainty around the apparently optimal approach. Cost-effectiveness models also illustrate the trade-offs inherent in each potential alternative available to the decision-maker and should be updated as new evidence becomes available.
10.1111/j.1524-4733.2006.00112.x
In 2000, The Academy of Managed Care Pharmacy issued its first Format for Formulary Submissions, which provided US payers with a new tool to proactively request clinical and economic data from health technology manufacturers. The Format, which has been revised substantially in recent years, is now used by health-care organizations covering some 150 million lives in the United States [1]. This guidance recommends that dossiers for new drug products be requested by health plans approximately 6 months before launch, and explicitly calls for the use of economic models “to inform decisions about the value or cost-effectiveness of pharmaceuticals, biologics, and vaccines.” Such models are to be based on