HOPE-3

Published

Yusuf S, et al. "Blood-Pressure and Cholesterol Lowering in Persons without Cardiovascular Disease". The New England Journal of Medicine. 2016. 374(21):2032-2343.
PubMed • Full text • PDF

Clinical Question

Among patients with intermediate cardiovascular disease risk, does rosuvastatin, hydrochlorthiazide/candesartan, or a combination of both interventions reduce CVD events when compared to placebo?

Bottom Line

In a 'polypill'-style study that tested initiation of rosuvastatin, BP meds, both, or neither without specific cholesterol or BP targets among adults at intermediate risk for CVD events, statins reduced CVD events but BP medications did not.

Major Points

Note: Three interrelated HOPE-3 manuscripts were published simultaneously: Antihypertensives+statin treatment, statin treatment, and antihypertensive treatment. All three are reviewed here.

CVD is the leading cause of death in the US. Targeted BP control with antihypertensives and targeted cholesterol control with statins led to a decline in the number of CVD events through the 1990s. This individualized approach might be inefficient, however. Primary prevention trials of statins in higher risk populations (e.g., JUPITER, WOSCOP, CARDS) saw reduction is in CVD events. Whether such an approach would benefit lower risk adults was unclear. Further, adults with elevated BP (e.g., systolic BP ≥120 mm Hg) experience increased risk from CVD when compared to those with normal BP (e.g., systolic bp <120 mm Hg). Whether use of untitrated antihypertensive medications may further reduce CVD events was unclear. Broadly prescribing statins and antihypertensive medications without consideration for titration is a component of the "'polypill'" approach to population health. Here, large populations of non-high-risk adults are treated with well-tolerated primary prevention medications in a combination pill to further lower incident CVD events.

Published in April 2016, the HOPE-3 (Heart Outcomes Prevention Evaluation-3) trial was a 2x2 factorial design, multinational, double-blinded, placebo-controlled trial performed in an ethnically diverse population. It was conducted among patients with intermediate cardiovascular risk, defined as a 1% annual risk of a cardiovascular event without another indication for these medications. A total of 12,705 patients were randomized to rosuvastatin, hydrochlorothiazide+candesartan, all medications, or placebo only. Drug doses were fixed and not titrated to specific targets throughout the trial, though participant's care teams could initiate antihypertensives or statins outside of the trial protocol. With a median follow-up of 5.6 years, there was a reduced incidence of CVD outcomes for those receiving statins but not those receiving antihypertensives. A sub-group analysis in the BP-lowering study showed patients with higher baseline systolic blood pressures trended towards significant reductions in the CVD events.

HOPE-3 provides further evidence supporting broad use of statin therapy for primary prevention of CVD. It did not yet provide clear support for untitrated antihypertensive medication use in a polypill approach. In contrast, SPRINT (2015) found a clear benefit for titrated intensive BP control among high-risk adults.

Guidelines

As of May 2016, no guidelines have been published that reflect the results of this trial.

Design

• Multicenter, double-blind, 2x2 factorial design placebo-controlled trial
• N=12,705 patients with intermediate cardiovascular disease risk
  • Rosuvastatin + Candesartan/HCTZ (n=3,180)
  • Rosuvastatin + Placebo (n=3,181)
  • Candesartan/HCTZ + Placebo (n=3,176)
  • Double placebo (n=3,181)
• Follow-up: Median 5.6 years
• Analysis: Intention-to-treat
• Co-primary outcomes:
  • Composite of cardiovascular death, MI, stroke
  • Above outcome, plus resuscitated cardiac arrest, heart failure, revascularization

Population

Inclusion Criteria

• Women ≥65 years and men ≥55 years
• 1 or more CV risk factors
  • Waist/hip ratio ≥0.85 (women) or ≥0.90 (men)
  • Smoker or recent ex-smoker (last 5 years)
  • HDL-C < 1.3mmol/L (women) or < 1.0mmol/L (men)
  • Dysglycemia (IFG, IGT, uncomplicated diet controlled DM)
  • Early renal dysfunction (eGFR<60, Cr>1.4mg/dL, microalbuminuria)
    • excluded if hypertensive (BP > 130/80) or proteinuria present
  • Family history of early CHD (M<55y, F<65y)

Exclusion Criteria

• Documented, clinically manifest atherothrombotic CVD
• Any statin, ARB, ACE inhibitor, or thiazide indication
• Statin, ARB, or thiazide contraindication
• Chronic liver disease or high LFTs (3x ULN)
• Inflammatory muscle disease or high CK (3x ULN)
• Moderate renal dysfunction (eGFR<45 or Cr>2mg/dL)
• Fibrate or cyclosporine treatment
• Symptomatic hypotension
• Serious condition interfering with trial completion or participation
• Use of experimental pharmacological agent

Baseline Characteristics

From the candesartan-HCTZ + rosuvastatin arm.

• Demographics: Age 66 years, female sex 46%
  • Race/ethnicity: Chinese 29%, hispanic 27%, white 20%, south Asian 15%, other asian 5%, black 2%, other 2%
• CVD risk factors: Elevated waist:hip 87%, recent/current smoking 28%, low HDL 38%, impaired glucose tolerance 12%, "early" DM 6%, FH early CAD 26%, "early" CKD 3%, HTN 38%
  • 2 risk factors: 47%
  • ≥3 risk factors: 25%
• Anthropometrics: BP 138/82 mm Hg, HR 73 BPM, BMI 27, waist:hip 0.94
• Cholesterol: Total 201, LDL 127, HDL 45, TG 128
• Fasting glucose: 96
• HS-CRP: 2.1 mg/L
• Creatinine: 0.90 mg/dL
• Medications: ASA 11%, BB 8%, CCB 14%

Interventions

• Eligible patients underwent a single-blind run-in phase receiving both cholesterol & BP-lowering agents for 4 weeks, and were excluded for poor adherence or unacceptable adverse events
• Randomization in 2 by 2 factorial design:
  • Randomization A: Lipid-lowering Arm vs. Placebo
  • Randomization B: BP-Lowering Arm vs. Placebo
• Arm-Specific Interventions
  • Lipid-lowering Arm: rosuvastatin 10mg daily
  • BP-lowering arm: candesartan 16mg daily and hydrochlorothiazide 12.5mg daily
• General Interventions
  • Individualized structured lifestyle advice
• Follow-up
  • Visits at 6 weeks, 6 months, then every 6 months thereafter
  • Blood pressure measured at every visit in the first year, then annually
  • Lipid levels at 1 year, 3 years and end of trial in an ethnically representative subsample of 10-20% of participants
• Open-label statins or antihypertensives could be initiated

Outcomes

Comparisons are intervention (reference) vs. placebo.

Primary Outcomes

P values of significance were adjusted to p=0.04 and p=0.02 for the below outcomes respectively, to preserve an overall type I error rate of 5% in testing two outcomes in each study.

Cardiovascular death, MI, or stroke

Statin study: 3.7% vs. 4.8% (HR 0.76; 95% CI 0.64-0.91; P=0.002)

BP-lowering study: 4.1% vs. 4.4% (HR 0.93; 95% CI 0.79-1.10; P=0.40)

Statin/BP-lowering study: 3.6% vs. 4.6% (HR 0.71; 95% CI 0.56-0.90; P=0.001)

Above outcome, plus resuscitated cardiac arrest, heart failure, revascularization

Statin study: 4.4% vs. 5.7% (HR 0.75; 95% CI 0.64-0.88; P<0.001)

BP-lowering study: 4.9% vs. 5.2% (HR 0.95; 95% CI 0.81-1.11; P=0.51)

Statin/BP-lowering study: 4.3% vs. 5.9% (HR 0.72; 95% CI 0.57-0.89; P=0.001)

Secondary Outcomes

All primary outcomes, plus angina with evidence of ischemia

Statin study: 4.8% vs. 6.2% (HR 0.77; 95% CI 0.66-0.89; P<0.001)

BP-lowering study: 5.3% vs. 5.7% (HR 0.92; 95% CI 0.79-1.06; p=0.26)

Statin/BP-lowering study: 4.6% vs. 6.5% (HR 0.71; 95% CI 0.57-0.87; P=0.001)

Fatal or nonfatal stroke

BP-lowering study: 1.2% vs. 1.5% (HR 0.80; 95% CI 0.59-1.08; P=0.14)

Primary/Secondary Outcomes, Components

Cardiovascular death

Statin study: 2.4% vs. 2.7% (HR 0.89; 95% CI 0.72-1.11)

Bp-lowering study: 2.4% vs. 2.7% (HR 0.91; 95% CI 0.73-1.13)

Statin/BP study: 2.4% vs. 2.9% (HR 0.82, 95% CI 0.60-1.11)

All MIs

Statin study: 0.7% vs. 1.1% (HR 0.65; 95% CI 0.44-0.94)

Bp-lowering study: 0.8% vs. 1.0% (HR 0.84; 95% CI 0.58-1.21)

Statin/BP study: 0.7% vs. 1.2% (HR 0.55; 95% CI 0.32-0.93)

All strokes

Statin study: 1.1% vs. 1.6% (HR 0.70; 95% CI 0.52-0.95)

BP-lowering study: 1.2% vs. 1.5% (HR 0.80; 95% CI 0.59-1.08)

This was a secondary endpoint, here for comparison purposes

Statin/BP study: 1.0% vs. 1.7% (HR 0.56; 95% CI 0.36-0.87)

Resuscitated Cardiac Arrest

Statin study: 0.1% vs. 0.1%

Bp-lowering study: <0.1% vs. 0.1% (HR 0.33, 95% CI 0.07-1.65)

Statin/BP study: <0.1% vs. 0.1% (HR 0.33, 95% CI 0.03-3.18)

Revascularization

Statin study: 0.9% vs.1.3% (HR 0.68; 95% CI 0.48-0.95)

Bp-lowering study: 1.0% vs. 1.2% (HR 0.86, 95% CI 0.62-1.21)

Statin/BP study: 0.8% vs. 1.4% (HR 0.59; 95% CI 0.37-0.95)

Heart Failure

Statin study: 0.3% vs. 0.5% (HR 0.72; 95% CI 0.41-1.26)

Bp-lowering study: 0.3% vs. 0.5% (HR 0.72; 95% CI 0.41-1.27)

Statin/BP study: 0.3% vs. 0.6% (HR 0.55; 95% CI 0.25-1.19)

Angina with objective evidence of ischemia

Statin study: 0.9% vs. 1.0% (HR 0.87; 95% CI 0.61-1.24)

Bp-lowering study: 0.8% vs. 1.1% (HR 0.74; 95% CI 0.51-1.06)

Statin/BP study: 0.8% vs. 1.2% (HR 0.65; 95% CI 0.39-1.08)

Subgroup Analyses

Statin Study

Benefits remained consistent across baseline cardiovascular risk, LDL, BP, and CRP levels

No heterogeneity by sex, age, race or ethnic group

BP Study (By level of baseline SBP)

Trend towards co-primary outcome (P=0.02), second co-primary outcome (P=0.009), first secondary outcome (P=0.005)

No trend towards stroke differences (P=0.22)

Patients in the upper third of baseline SBP, compared to placebo had a nominally significant reduction of the first coprimary outcome (HR 0.73), secondary coprimary outcome (HR 0.76), and first secondary outcome (HR 0.72)

No interactions among baseline risk, LDL, diastolic BP

Combination Study

No significant interactions by stroke risk, LDL, sBP

No heterogeneity in effects by age, sex, race or ethnic group

Statin-only vs. BP-lowering arm: nonsignificant trend towards lower risk of first coprimary outcome (3.8% vs. 4.6%, HR 0.82; 95% CI 0.65-1.05, P=0.11), and marginally significant for second coprimary outcome (4.4% vs. 5.5%, HR 0.79, 95% CI 0.64-0.99, P=0.04)

Additional outcomes

Total # of second primary outcome events

169 vs. 262 events (HR 0.66; 95% CI 0.52-0.84; P=0.001)

Hospitalizations from cardiovascular causes

4.4% vs. 6.0% (HR 0.73; 95% CI 0.59-0.91; P=0.005)

There was no significant difference between all-cause mortality, new-onset diabetes, or hospitalization from non-cardiovascular causes.

Adverse Events

BP Study

Discontinuation rates: 24.4% vs. 25.2%

Symptomatic hypotension, lightheadedness, dizziness: 3.4% vs. 2.1% (P<0.001)

Syncope: p=NS

Renal dysfunction: 0.5% vs. 0.3% (p=0.13)

Lipid Study

Discontinuation rates:

Discontinuation because of muscle symptoms: 1.3% vs. 1.2%

Muscle pain or weakness: 5.8% vs. 4.7% (p=0.005)

Rhabdomyolysis: 2 vs. 2

Myopathy: 1 vs. 1

Cataract surgery: 3.8% vs. 3.1% (p=0.02)

VTE: 14 vs. 31 (p=0.01)

Combination Study

Discontinuation rates: 26.3% vs. 28.8%

Criticisms

• Statins may have disproportionately benefited white participants and there may have been more statin myopathy among East Asians.
• CVD endpoints may have been driven by more revascularization events and nonfatal MIs.

Funding

• AstraZeneca (Atacand - Candesartan/HCTZ, Crestor - rosuvastatin) had a single voting member present on a 24-member steering committee
• Canadian Institutes of Health Research (CIHR)