Straight Talk: ACC 2016: Interpreting the Prevention Trials

CardioSource WorldNews | Steven E. Nissen, MD, MACC

The 2016 Annual Scientific Sessions of the American College of Cardiology provided an unusually impactful series of insights into the treatment of hyperlipidemia and hypertension. The long awaited HOPE-3 trial was finally reported, accompanied by an unprecedented three manuscripts in the New England Journal of Medicine. (1-3) The trial randomized 12,705 “intermediate risk” individuals to either low-dose rosuvastatin (10 mg daily), the combination of candesartan 16 mg plus hydrochlorothiazide 12.5 mg, or both therapies for a median follow-up of 5.6 years. The stated goal of the trial was to provide a “proof of concept” validation of the “polypill”, an often advocated approach to cardiovascular (CV) prevention that combines lose doses of several therapies into a single low-cost multi-drug combination product intended for administration to patients at high risk of CV morbidity/mortality. Some of the most ardent advocates of the polypill have suggested adopting this approach on a population-wide basis without customization to individual patient risk profiles.

The results of HOPE-3 clearly demonstrate the irrationality of the polypill concept. Although the rosuvastatin treatment group showed a statistically robust benefit (an approximately 25% reduction in death, stroke, and myocardial infarction), the blood pressure (BP) lowering arm yielded unimpressive results, a non-significant 7% reduction in the primary endpoint, p = 0.40. The authors emphasized the significant benefits in the patients within the upper tertile of BP (>143.5 mm Hg systolic), but didn’t emphasize the potential harm in patients within the lower tertile (≤131.5 mm Hg). Although the results for the lowest BP tertile didn’t reach statistical significance because of the limited statistical power, the hazard ratio (HR) for the broadest coprimary endpoint was 1.25, compared with 0.76 for the highest BP tertile. These data show that overly aggressive BP lowering in these patients is likely harmful rather than helpful. Accordingly, administering polypill therapy to a broad swath of the population would likely yield both benefits and hazards, depending on the patient’s risk profile. HOPE-3 reveals what most clinicians already know, individualization of treatment is best, rather than “one size fits all.”

Two other clinical trials generated important insights. We presented the GAUSS-3 Trial, simultaneously published in JAMA, that addressed the challenging clinical problem of statin intolerance due to muscle-related adverse effects. (4) The trial design was unusual, enrolling 511 patients with a strong history of intolerable muscle symptoms when administered multiple statins (82% failed 3 or more statins). These patients were initially randomized to atorvastatin or placebo for 10 weeks and asked to report whether they experienced intolerable muscle symptoms. Then, after a 2-week washout, patients crossed over to the alternative treatment group so that each patient had 10 weeks of statin and 10 weeks of placebo. Only patients with symptoms of atorvastatin, but not placebo, entered the second phase of the trial. During the blinded atorvastatin rechallenge, 42.6% of patients had intolerable muscle symptoms on the statin, but not placebo, while 26.5% had symptoms on placebo, but not atorvastatin.

These findings provide two key insights. First, muscle-related statin intolerance is a real phenomenon affecting a substantial fraction of patients. Second, there is a strong psychosomatic component of statin intolerance with about a quarter of patients having symptoms on placebo, a phenomenon sometimes called the “nocebo effect.” For the 42.6% of patients with confirmed statin intolerance, we subsequently randomized patients to receive ezetimibe or the new PCSK9 inhibitor, evolocumab. Not surprisingly, the evolocumab-treated patients had a much more robust lowering of low-density lipoprotein cholesterol (LDL-C), 52-54% compared with 16.7% for ezetimibe. Despite baseline LDL-C levels of about 220 mg/dL, approximately two thirds of evolocumab-treated patients achieved an LDL-C < 100 mg/dL, compared with less than 2% of ezetimibe-treated patients.

A third study presented at ACC 2016 stunned the lipidology community, the ACCELERATE Trial, presented by Stephen Nicholls, MD, FACC. This study was presented prior to final closeout, but represented nearly 99% complete data. This was a 12,092 patient trial that randomized statin-treated patients at high risk for cardiovascular events to a cholesteryl ester transfer protein (CETP) inhibitor, evacetrapib, or placebo. This drug yielded lipid-modulating benefits exactly as expected, a 130% increase in high-density lipoprotein cholesterol (HDL-C) and a 37% decrease in LDL-C. Despite these large lipid benefits, the study was terminated for futility on the recommendation of the Data Monitoring Committee after a median study participation of 29 months. The primary endpoint was a composite of cardiovascular death, MI, stroke, coronary revascularization, or hospitalization for unstable angina. A total of 774 events occurred in the evacetrapib group and 768 events in the placebo group, HR 1.01, p = 0.85. None of the components of the primary endpoint showed statistically significant benefits, although there was a trend toward benefit for all-cause mortality, p = 0.06. A small increase in BP (0.9 mm) and C-reactive protein (4.6%) do not likely explain the lack of benefit.

The results represent the third CETP inhibitor to fail to show evidence of meaningful clinical benefits (torcetrapib, dalcetrapib, and now evacetrapib). These findings represent a stunning failure of favorable lipid effects to translate into an improvement in clinical outcomes. Once again, the use of surrogate endpoints like cholesterol levels, reduction in glycemia, or BP lowering, have proven unreliable in predicting reductions in the intended cardiovascular benefits. ACCELERATE should serve as a warning and a reminder that excessive faith in biochemical “benefits” has frequently mislead the cardiovascular community. Despite the high costs and time-consuming nature of clinical outcome trials, large, carefully conducted randomized controlled trials represent the only reliable approach to determining whether therapies are beneficial, neutral, or harmful.


  1. Lonn EM, Bosch J, López-Jaramillo P, et al. N Engl J Med. 2016 Apr. [Epub ahead of print]
  2. Yusuf S, Bosch J, Dagenais G, et al. N Engl J Med. 2016 Apr 2. [Epub ahead of print]
  3. Yusuf S, Lonn E, Pais P, et al. N Engl J Med. 2016 Apr 2. [Epub ahead of print]
  4. Nissen SE, Stroes E, Dent-Acosta RE, et al. JAMA. 2016 Apr 3. doi: 10.1001/jama.2016.3608. [Epub ahead of print]

Steven E. Nissen, MD, is Chair of the Department of Cardiovascular Medicine at the Cleveland Clinic and co-author of Heart411: The Only Guide to Heart Health You’ll Ever Need.

Read the full May issue of CardioSource WorldNews at

Clinical Topics: Dyslipidemia, Prevention, Novel Agents, Statins, Hypertension

Keywords: CardioSource WorldNews, Benzimidazoles, Drug Combinations, Hydrochlorothiazide, Hyperlipidemias, Hypertension, Tetrazoles

< Back to Listings