You are here
Variations in time to benefit among clinical trials of cholesterol-lowering drugs
J Clin Lipidol. 2018;12(4):857-862
Time to benefit (TTB) in clinical trials of cholesterol-lowering drugs is important because it may provide a clue as to the potential mechanism of action of the drug, it is helpful in determining when to stop a trial for futility, and it may inform treatment decisions in subjects with reduced life expectancy.
To compare TTB among clinical trials of cholesterol-lowering drugs.
We examined TTB in 24 trials of cholesterol-lowering drugs with positive outcomes. Benefit curves were constructed by subtracting the curve for a placebo or comparator drug from the curve for active treatment.
TTB ranged from 1 to 30 (mean 13.1) months, being shorter in trials of statins (n = 17) compared to nonstatins (n = 7), 10.3 vs 20.0 months. Among statin trials, TTB was shorter with atorvastatin (n = 6) than in trials with other statins (n = 11), 4.75 compared to 11.4 months.
TTB is variable among trials of cholesterol-lowering drugs, being shorter with statin compared to nonstatin drugs. TTB is shorter with atorvastatin than with other statins. For trials of new cholesterol-lowering drugs, outcome curves that do not separate for up to 30 months do not preclude eventual benefit.
Time to benefit (TTB) is quite variable across trials of lipid-lowering drugs.
TTB appears to be shorter with statins than with other drugs.
Among statin trials, TTB is shorter with atorvastatin than with other statins.
For new trials, curves that do not separate for 24-30 months do not preclude eventual benefit.
Keywords: Cholesterol lowering, Clinical trials, Statins, Time to benefit.
Since the dawn of the modern era in clinical trials, outcome curves have attracted detailed scrutiny. After an early separation, do they continue to widen or do they run parallel? If one treatment is associated with early hazard, such as coronary bypass surgery, how long does it take for the curves to cross? And do the curves eventually converge after a number of years?
Arguably, the center of interest for outcome curves is the point where they diverge, here considered to be equivalent to time to benefit (TTB). TTB attracts attention for at least 3 reasons. First, it may provide insight into the mechanism through which a treatment yields benefit. For example, if a low-density lipoprotein cholesterol (LDL-C)–lowering drug begins to reduce cardiovascular (CV) events before significant LDL-C lowering has occurred, it suggests that perhaps a mechanism other than LDL-C lowering may account for the benefit. 1 Second, TTB is an important consideration in deciding whether or not, and when, to stop a clinical trial because of futility. Figure 1 depicts the outcome curves for 3 trials of cholesterol-lowering drugs where the curves did not separate appreciably until after 2-3 years but where statistically significant benefit was eventually demonstrated. Third, it has been advocated that TTB should be taken into account when deciding whether to treat elderly patients with multiple diseases and reduced life expectancy. 5 Treatment would be futile if TTB was longer than the patient's lifespan.
TTB in LDL-C-lowering trials
More than 2 dozen large, randomized, controlled, double blind, CV clinical outcome trials with statins have been completed; most of them achieved their primary endpoint. In some, the comparator to the statin was a placebo, 2 4 6 7 8 9 10 11 12 13 14 15 16 17 whereas in others, different statins or different doses of the same statin were compared. 18 19 20 In addition, several studies with positive outcomes by using other drugs that lower LDL-C have been performed (cholestyramine, gemfibrozil, ezetimibe, evolocumab, alirocumab, and anacetrapib). 3 21 22 23 24 25 26
The salient features of these trials are listed in Table 1 . TTB was determined by visual inspection of the outcome curves for the primary endpoint. TTB varied from 1 to 36 (mean 13.1) months across all 24 trials, averaging 11.1 months in statin vs placebo trials (n = 14), 10.3 months in all statin trials (n = 17), and 20.0 months in nonstatin trials (n = 7). TTB was 12 months in each of the 2 trials of PCSK9 inhibitors.
|Trial||Drug and statin dose (mg/d)||Comparator||Number in trial||Number of events||Baseline LDL-C (mg/dL)||LDL-C reduction (mg/dL)||Primary endpoint reduction (%)||Trial duration (y)||Estimated TTB (mo)|
|Statin vs Placebo|
|4S 6||Simvastatin 20/40||Placebo||4444||855||189||64||34||5.4||12|
|JUPITER 7||Rosuvastatin 20||Placebo||17,802||393||108||54||44||1.9||3|
|MIRACL 8||Atorvastatin 80||Placebo||3086||497||124||52||16||0.3||1.5|
|WOSCOPS 9||Pravastatin 40||Placebo||6595||422||192||50||31||4.9||6|
|CARDS 10||Atorvastatin 10||Placebo||2838||210||116||46||37||3.9||6|
|HPS 11||Simvastatin 40||Placebo||20,536||2110||131||43||24||5.0||12|
|AFCAPS 12||Lovastatin 20/40||Placebo||6605||299||150||38||37||5.2||6|
|ASCOT-LLA 13||Atorvastatin 10||Placebo||10,305||254||131||37||36||3.3||1|
|LIPID 14||Pravastatin 40||Placebo||9014||299||150||36||24||6.0||18|
|HOPE-3 4||Rosuvastatin 10||Placebo||12,705||539||128||35||24||5.6||30|
|CARE 2||Pravastatin 40||Placebo||4159||486||139||35||24||5.0||24|
|PROSPER 15||Pravastatin 40||Placebo||5804||881||147||50||15||3.2||18|
|MEGA 16||Pravastatin 10-20||None – open label||7832||167||156||23||33||5.3||6|
|SHARP 17||Ezetimibe/Simvastatin 10/20||Placebo||9270||1145||107||33||17||4.9||12|
|High vs low dose|
|PROVE-IT 18||Atorvastatin 80||Pravastatin 40||4162||1012||95||33||16||2.5||1|
|TNT 19||Atorvastatin 80||Atorvastatin 20||10,001||982||101||24||22||5.5||1|
|IDEAL ∗ , 20||Atorvastatin 80||Simvastatin 20/40||8888||874||121||23||11||4.8||18|
|IMPROVE-IT 23||Ezetimibe/Simvastatin 10/20||Simvastatin||18,144||5314||69||16||6.4||7.4||14|
|REVEAL 26||Anacetrapib||Placebo||30,449||3443||61||26 †||9||4.1||30|
∗ The IDEAL trial was negative for the primary endpoint ( P = .07). In this analysis, stroke is added to the primary endpoint so that it is identical to the TNT endpoint.
† This number is an overestimate of the percentage LDL-C lowering because of problems with assaying LDL-C when cholesteryl ester transfer protein is inhibited. In a subset of patients where LDL-C was measured using beta quantitation, the reduction was 17%.LDL-C, low-density lipoprotein cholesterol; TTB, time to benefit.
TTB in a selection of these trials is shown in Figure 2 , where for the primary outcomes, the comparator curve has been subtracted from the curve for active treatment. The resulting curve thus begins at TTB on the x-axis and represents the additional benefit of active treatment over the comparator. This analysis is an approximation because it is not derived from patient data but from manipulation of the published outcome curves. To avoid clutter, only 15 of the 24 trials included in Table 1 are shown.
Factors potentially influencing differences in TTB
Several factors that might explain the differences in TTB among the trials are listed in Table 2 . Obviously, benefit will not appear in a treatment group until at least a few endpoint events have occurred. The number of endpoint events depends on the number of patients at risk (sample size) and the event rate. A trial with 200 events within the first 6 months will be more likely to show early benefit than a trial with only 20 events during the same interval.
The amount of LDL-C reduction correlates with the magnitude of CV event reduction; specifically, each mmol/L (38.6 mg/dL) reduction in LDL-C in statin trials corresponds to a 22% reduction in events. 27 In statin vs nonstatin trials, the reduction in CV events is 9% in the first year and 22%-28% in subsequent years, per mmol/L reduction in LDL-C. 27 In general, a more robust reduction in events would be expected to shorten TTB. However, if one plots the magnitude of LDL-C reduction vs TTB using the data in Table 1 , the correlation is poor.
Baseline LDL-C is another factor with the potential to affect TTB. For a given percentage reduction in LDL-C, the absolute reduction in LDL-C will increase with increasing baseline levels such that higher baseline levels are associated with greater absolute LDL-C reduction. Very high baseline LDL-C levels were present in only 2 trials, the Lipid Research Clinics Coronary Primary Prevention Trial 21 and the Scandinavian Simvastatin Survival Study (4S) 6 ; in neither was TTB particularly short. After 4S, it was no longer considered ethical to leave subjects with such high levels of LDL-C untreated. Considering all the trials with positive outcomes, TTB was unrelated to the baseline level of LDL-C.
The type of patients enrolled in a trial might also affect TTB. Patients with an acute coronary syndrome (ACS) experience a much higher event rate over the next 6 months when compared to stable coronary disease patients; the greater number of early events might be predicted to shorten TTB. Statins have an early beneficial effect on a variety of harmful pathophysiologic mechanisms active in ACS patients: endothelial dysfunction, inflammation, platelet reactivity, and hypercoagulability. 1 28 It is now widely accepted that statins have an early beneficial effect on CV events in these patients due to these mechanisms. The use of ezetimibe in a large post-ACS population (IMPROVE-IT) was not accompanied by a short TTB. 23
In the absence of ACS, TTB appears to be similar in patients both without known coronary disease and in patients with stable coronary disease. For the primary prevention studies listed in Table 1 , TTB covers a very wide range, from 1 13 to 30 4 months.
Does TTB depend on the specific drug used to lower LDL-C? The evidence suggests that it does. From Table 1 , the average TTB in the 17 statin trials was 10.3 months, compared to 20.0 months in the 7 trials with other drugs. Among the statins, atorvastatin appears to be associated with a shorter TTB compared to the others. The average TTB in the 6 atorvastatin trials was 4.75 months, compared to 13.4 months in the 11 trials with other statins.
Is there an explanation for the shorter TTB with atorvastatin? The factors listed in Table 2 and discussed previously do not account for the difference. Some evidence suggests that atorvastatin has an active metabolite that functions as an antioxidant and has a favorable effect on lipoproteins. 29 30 31 A comparable metabolite has not been observed with other statins. It must be stressed, however, that a suggestion that the antioxidant metabolite of atorvastatin is the explanation for the shorter TTB in atorvastatin trials is speculative. It is possible that the shorter TTB in the atorvastatin trials is the consequence of an unknown mechanism, or that it is due to random variation among the trials.
The results of this analysis should be of interest to those who design or evaluate the design of clinical trials of cholesterol-lowering drugs. Although some trials, particularly those with atorvastatin, show benefit within the first year, early benefit should not be expected with all LDL-C-lowering drugs. In fact, an absence of any outcome curve separation up to 30 months does not preclude eventual benefit. Stopping a trial early for futility might consign an effective drug to oblivion, with the loss of drug development costs. A failed trial also impacts on the development of other new drugs from the same class. A false-negative trial also misleads the whole field in the search for drug targets and an understanding of the pathophysiology.
The assessment of TTB as conducted in this analysis lacks statistical validity. At the point the curves separate, the difference between them does not approach statistical significance in most instances, and with rare exceptions, 1 authors do not calculate when benefit attains statistical significance. In fact, when the difference between treatments does become statistically significant, the trial may be stopped early. 10 13 In addition, determining when the curves separate is an estimation, associated with both a degree of error and an unknown amount of interobserver variability. One major journal now displays outcome curves in 2 formats, one with a y-axis stretching from 0% to 100% and an inset with a truncated y-axis showing only the area of interest. In at least one trial included in this analysis, where the difference between the groups is marginal anyway, TTB looks much different in the different formats. 23
This report is limited to the topic of TTB and does not address other important aspects of clinical trials. For example, duration of benefit is also important. Long-term follow-up of patients in statin trials indicates that benefit persists and even increases after the trial ends, when statin use equalizes in the original treatment groups. 32 On the other hand, the long-term benefit of beta-blockers after myocardial infarction in the absence of heart failure is questionable because the rate of metabolic adverse effects such as new-onset diabetes increases and benefit may be attenuated. 33 The clinical features and concomitant treatments of myocardial infarction survivors have also changed dramatically over recent decades, such that the results of older trials may no longer apply. 33 Extrapolating the results of a clinical trial to a different drug or dose, or to a different patient population, may lead to error.
In conclusion, TTB varies from 1 to 30 months across trials of lipid-lowering drugs. Baseline LDL-C and the magnitude of LDL-C lowering do not correlate with TTB. TTB may be shorter in ACS trials using a statin, perhaps due to statin benefits unrelated to LDL-C lowering. TTB is shorter with statins compared to other lipid-lowering drugs. TTB is shorter with atorvastatin than with other statins. In trials with new lipid-lowering drugs, patience should be exercised before declaring futility, as benefit may develop only after 2 years.
Authors' contributions: Both authors contributed to study design, data analysis, manuscript preparation, and revision. Both authors have approved the final version and submission to this journal. The authors wish to thank Shirley Smith, PhD, from Engage Scientific, Envision Pharma Group, U.K., for creating Figure 2 .
D.D.W. received remuneration for participating in clinical trial committees CSL Ltd, Pfizer, Regeneron, Resverlogix, and Sanofi. D.D.W. received honoraria for lectures and consulting fees from Pfizer. P.J.B. received grant support from Pfizer and Merck; fees for serving on an advisory board and lecture fees from Pfizer, Merck, and Amgen; and fees for serving on an advisory board from AstraZeneca and Sanofi-Regeneron.
- 1 K.K. Ray, C.P. Cannon, C.H. McCabe, et al. Early and late benefits of high-dose atorvastatin in patients with acute coronary syndromes: results from the PROVE IT-TIMI 22 trial. J Am Coll Cardiol. 2005;46:1405-1410
- 2 F.M. Sacks, M.A. Pfeffer, L.A. Moye, et al. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med. 1996;335:1001-1009
- 3 H.B. Rubins, S.J. Robins, D. Collins, et al. Gemfibrozil for secondary prevention of coronary heart disease in men with low levels of high-density lipoprotein cholesterol. N Engl J Med. 1999;341:410-418
- 4 S. Yusuf, J. Bosch, G. Dagenais, et al. Cholesterol lowering in intermediate-risk persons without cardiovascular disease. N Engl J Med. 2016;374:2021-2031
- 5 H.M. Holmes, L.C. Min, M. Yee, et al. Rationalizing prescribing for older patients with multimorbidity; considering time to benefit. Drugs Aging. 2013;30:655-666
- 6 Scandinavian Simvastatin Survival Study Group. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994;344:1383-1389
- 7 P.M. Ridker, E. Danielson, F.A.H. Fonseca, et al. Rosuvastatin to prevent vascular events in men and women with elevated C-reactive protein. N Engl J Med. 2008;359:2195-2207
- 8 G.G. Schwartz, A.G. Olsson, M.D. Ezekowitz, et al. Effect of atorvastatin on early recurrent ischemic events in acute coronary syndromes. The MIRACL study: a randomized controlled trial. JAMA. 2001;285:1711-1718
- 9 J. Shepherd, S.M. Cobbe, I. Ford, et al. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. N Engl J Med. 1995;333:1711-1718
- 10 H.M. Colhoun, D.J. Betteridge, P.N. Durrington, et al. Primary prevention of cardiovascular disease with atorvastatin in type 2 diabetes in the Collaborative Atorvastatin Diabetes Study (CARDS): multicenter randomised placebo-controlled trial. Lancet. 2004;364:685-696
- 11 Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20536 high-risk individuals: a randomised placebo-controlled trial. Lancet. 2002;360:7-22
- 12 J.R. Downs, M. Clearfield, S. Weiss, et al. Primary prevention of acute coronary events with lovastatin in men and women with average cholesterol levels. Results of AFCAPS/TexCAPS. JAMA. 1998;279:1615-1622
- 13 P.S. Sever, B. Dalhöf, N.R. Poulter, et al. Prevention of coronary and stroke events with atorvastatin in hypertensive patients who have average or lower-than-average cholesterol concentrations, in the Anglo-Scandinavian Cardiac Outcomes Trial - Lipid Lowering Arm (ASCOT-LLA): a multicenter randomised controlled trial. Lancet. 2003;361:1149-1158
- 14 The Long-Term Intervention with Pravastatin in Ischaemic Heart Disease (LIPID) Study Group. Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels. N Engl J Med. 1998;339:1349-1357
- 15 J. Shepherd, G.J. Blauw, M.B. Murphy, et al. Pravastatin in elderly individuals at risk of vascular disease (PROSPER): a randomized controlled trial. Lancet. 2002;360:1623-1630
- 16 H. Nakamura, K. Arakawa, H. Itakura, et al. Primary prevention of cardiovascular disease with pravastatin in Japan (MEGA Study): a prospective randomised controlled trial. Lancet. 2006;368:1155-1163
- 17 C. Baigent, M.J. Landray, C. Reith, et al. The effects of lowering LDL cholesterol with simvastatin plus ezetimibe in patients with chronic kidney disease (Study of Heart and Renal Protection): a randomized placebo-controlled trial. Lancet. 2011;377:2181-2192
- 18 C.P. Cannon, E. Braunwald, C.H. McCabe, et al. Intensive versus moderate lipid lowering with statins in acute coronary syndromes. N Engl J Med. 2004;350:1495-1504
- 19 J.C. LaRosa, S.M. Grundy, D.D. Waters, et al. Intensive lipid lowering with atorvastatin in patients with stable coronary disease. N Engl J Med. 2005;352:1425-1435
- 20 T.R. Pedersen, O. Faergeman, J.J.P. Kastelein, et al. High-dose atorvastatin vs usual-dose simvastatin for secondary prevention after myocardial infarction. The IDEAL study: a randomized controlled trial. JAMA. 2005;294:2437-2445
- 21 Lipid Research Clinics Program. The Lipid Research Clinics Coronary Primary Prevention Trial results. JAMA. 1984;251:35-64
- 22 M.H. Frick, O. Elo, K. Haapa, et al. Helsinki Heart Study: Primary prevention trial with gemfibrozil in middle-aged men with dyslipidemia. N Engl J Med. 1987;317:1237-1245
- 23 C.P. Cannon, M.A. Blazing, R.P. Giugliano, et al. Ezetimibe added to statin therapy for acute coronary syndromes. N Engl J Med. 2015;372:2387-2397
- 24 M.S. Sabatine, R.P. Giugliano, A.C. Keech, et al. Evolocumab and clinical outcomes in patients with cardiovascular disease. N Engl J Med. 2017;376:1713-1722
- 25 P.G. Steg, G.G. Schwartz, M. Szarek, et al. ODYSSEY Outcomes trial, presented March 10, 2018, at the American College of Cardiology Annual Scientific Sessions. http://www.acc.org/latest-in-cardiology/articles/2018/03/05/15/53/sat-9am-odyssey-outcomes-cv-outcomes-with-alirocumab-after-acs-acc-2018 Available at:
- 26 The HPS3/TIMI55–REVEAL Collaborative Group. Effects of anacetrapib in patients with atherosclerotic vascular disease. N Engl J Med. 2017;377:1217-1227
- 27 R. Collins, C. Reith, J. Emberson, et al. Interpretation of the evidence for the efficacy and safety of statin therapy. Lancet. 2016;388:2532-2561
- 28 D.D. Waters. Cholesterol lowering: should it continue to be the last thing we do?. Circulation. 1999;99:3215-3217
- 29 R.P. Mason, M.F. Walter, C.A. Day, et al. Active metabolite of atorvastatin inhibits membrane cholesterol domain formation by an antioxidant mechanism. J Biol Chem. 2006;281:9337-9345
- 30 R.F. Jacob, M.F. Walter, Y. Self-Medlin, et al. Atorvastatin active metabolite inhibits oxidative modification of small dense low-density lipoprotein. J Cardiovasc Pharmacol. 2013;62:160-166
- 31 R.P. Mason, S.C.R. Sherratt, R.F. Jacob. Eicosapentaenoic acid inhibits oxidation of apoB-containing lipoprotein particles of different size in vitro when administered alone or in combination with atorvastatin active metabolite compared with other triglyceride-lowering agents. J Cardiovasc Pharmacol. 2016;68:33-40
- 32 I. Ford, H. Murray, C. McCowan, C.J. Packard. Long-term safety and efficacy of lowering low-density lipoprotein cholesterol with statin therapy. 20-year follow-up of West of Scotland coronary prevention study. Circulation. 2016;133:1073-1080
- 33 D.D. Waters, S. Bangalore. The evolution of myocardial infarction: when the truths we hold to be self-evident no longer have evidence. Can J Cardiol. 2017;33:1209-1211
© 2018 National Lipid Association. All rights reserved.
© 2018 National Lipid Association. All rights reserved.
Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds or experiments described herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. To the fullest extent of the law, no responsibility is assumed by Elsevier or the National Lipid Association for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein.
This e-print is distributed with the support of Pfizer.
Elsevier España, S.L.U.
(A member of Elsevier)
Av. Josep Tarradellas, 20-30
Tel.: 932 000 711
Fax: 932 091 136