The Lancet NOVEMBER 23, 2002

Prospective Study of Pravastatin in the Elderly at Risk (PROSPER)

Shepherd J, Blauw GJ, Murphy MB, et al. (for the PROSPER study group)

Bottom Line

The PROSPER trial demonstrated that daily treatment with 40 mg of pravastatin significantly reduces the risk of a composite cardiovascular endpoint in high-risk patients aged 70 to 82 years, although it showed no effect on stroke, cognitive function, or disability.

Key Findings

1. Pravastatin treatment resulted in a 15% relative risk reduction in the primary composite endpoint of coronary heart disease death, non-fatal myocardial infarction, and fatal or non-fatal stroke (14.1% in the pravastatin group vs. 16.2% in the placebo group; hazard ratio 0.85, 95% CI 0.74–0.97, p=0.014).

2. Secondary endpoints showed that coronary events were reduced by 19% (HR 0.81, p=0.006) and coronary mortality was reduced by 24% (HR 0.76, p=0.043) in the pravastatin arm.

3. There was no observed clinical benefit regarding the prevention of stroke (4.7% with pravastatin vs. 4.5% with placebo, p=0.81).

4. Pravastatin therapy did not demonstrate any treatment effect on heart failure requiring hospitalization, disability, or cognitive function.

5. A higher incidence of new cancer diagnoses was observed in the pravastatin group (8.5% vs 6.8%, HR 1.25, 95% CI 1.04–1.51, p=0.020), though this finding was not supported by subsequent meta-analyses of broader statin trial data.

Study Design

Design

RCT

Double-Blind

Sample

5,804

Patients

Duration

3.2 yr

Median

Setting

Multicenter, Europe

Population Men and women aged 70–82 years with pre-existing vascular disease or at high risk for vascular disease due to smoking, hypertension, or diabetes, with total cholesterol between 4.0 and 9.0 mmol/L.

Intervention Pravastatin 40 mg daily

Comparator Matching placebo

Outcome Composite of coronary heart disease death, non-fatal myocardial infarction, and fatal or non-fatal stroke.

Study Limitations

• The study did not find significant benefits for non-cardiovascular outcomes, specifically stroke and cognitive function, despite original hypotheses suggesting potential neuroprotective effects.

• The observed increase in cancer incidence in the pravastatin group, while statistically significant, was likely a chance finding as it was not replicated in larger meta-analyses.

• The findings are restricted to patients aged 70–82 years with specific lipid profiles, potentially limiting generalizability to the 'oldest-old' population (e.g., >85 years) or those with different baseline cholesterol levels.

• The trial was conducted in a specific geographic region (Scotland, Ireland, and the Netherlands) which may influence the applicability of findings to more diverse demographic settings.

Clinical Significance

PROSPER provided foundational evidence that cholesterol-lowering therapy with statins is effective for cardiovascular risk reduction even in patients over the age of 70, effectively extending established primary and secondary prevention strategies from middle-aged cohorts to the elderly population.

Historical Context

Prior to PROSPER, most large-scale statin trials (such as 4S, LIPID, and HPS) focused on middle-aged participants. PROSPER was the first major randomized controlled trial designed specifically to evaluate the efficacy and safety of statin therapy in an elderly population, addressing a critical gap in evidence-based cardiovascular management for older adults.

Guided Discussion

High-yield insights from every perspective

Med Student

Medical Student

Pravastatin works by inhibiting HMG-CoA reductase. In the context of the PROSPER trial's elderly population, why was the assessment of cognitive function and disability just as clinically relevant as the assessment of cardiovascular death?

Key Response

In geriatric medicine, the 'quality of life' and 'functional independence' are often prioritized alongside survival. Since cholesterol is a component of neuronal membranes and vascular health is linked to dementia (vascular dementia), it was hypothesized that statins might either protect cognitive function via vascular stabilization or impair it through cholesterol depletion. PROSPER demonstrated that while cardiac events decreased, there was no cognitive benefit or harm.

Resident

A 78-year-old patient with hypertension and smoking history asks if they should start pravastatin based on the PROSPER trial. While the trial showed a reduction in the primary composite endpoint, what specific 'negative' findings regarding stroke and total mortality should you discuss with the patient to manage expectations?

Key Response

Residents must recognize that PROSPER showed a reduction in coronary heart disease (CHD) death and non-fatal MI, but failed to show a significant reduction in stroke or all-cause mortality. This is a critical distinction for shared decision-making, as it suggests that in the short term (3.2 years), statins in the elderly primarily prevent heart attacks rather than extending overall lifespan or preventing strokes.

Fellow

PROSPER is often cited for the 'stroke paradox'—a reduction in coronary events without a corresponding reduction in stroke. How does the pathophysiology of cerebrovascular disease in the 70-82 age bracket potentially explain this discrepancy compared to younger cohorts in trials like 4S or WOSCOPS?

Key Response

In the very elderly, strokes are more frequently cardioembolic (e.g., from atrial fibrillation) or related to small-vessel disease and amyloid angiopathy, whereas coronary events remain predominantly atherosclerotic. Statins target plaque stabilization and LDL reduction, which may have a more immediate impact on large-vessel coronary beds than on the multifactorial stroke mechanisms prevalent in octogenarians over a short 3-year horizon.

Attending

The PROSPER trial noted a statistically significant increase in new cancer diagnoses in the pravastatin group (HR 1.25). How should an attending physician teach trainees to interpret this 'signal' in light of subsequent meta-analyses such as those by the Cholesterol Treatment Trialists' (CTT) Collaboration?

Key Response

This is a classic teaching point on 'trial-specific findings' versus 'class effects.' While PROSPER raised an alarm about cancer, subsequent meta-analyses of over 170,000 participants showed no association between statins and cancer incidence. Attending physicians use this to teach the importance of not over-interpreting secondary safety signals from a single trial and instead relying on the totality of evidence.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD

The PROSPER trial utilized a PROBE (Prospective Probit Open Blinded End-point) design. Evaluate the internal validity threats this design poses for the secondary endpoints of MMSE (Mini-Mental State Examination) and ADL (Activities of Daily Living) scores compared to the primary hard cardiovascular endpoints.

Key Response

While hard endpoints like death or MI are less susceptible to observer bias, functional and cognitive assessments (MMSE/ADL) are more subjective. In an open-label design (even with blinded end-point adjudication), the knowledge of treatment assignment by the patient or the performing clinician could lead to subtle differences in effort or reporting, potentially masking or exaggerating the drug's effect on quality-of-life metrics.

Journal Editor

If you were reviewing the PROSPER manuscript, how would you evaluate the clinical significance of the 3.2-year median follow-up period for a drug intended for primary prevention in a population with significant competing risks of mortality?

Key Response

A Journal Editor would flag that 3.2 years may be insufficient to see the full benefits of lipid-lowering, especially for endpoints like dementia or stroke, which have long incubation periods. Furthermore, in an elderly population, the 'competing risk' of non-cardiovascular death (like cancer or pneumonia) can dilute the observable benefit of a cardiovascular intervention, necessitating a rigorous survival analysis (e.g., Fine-Gray model) which was less standardized at the time of publication.

Guideline Committee

How does the PROSPER trial's evidence specifically inform the 2018 AHA/ACC Cholesterol Guidelines' Class IIb recommendation for statin initiation in primary prevention for adults >75 years old?

Key Response

PROSPER is a foundational trial for the elderly, but its failure to show a mortality benefit or stroke reduction led guideline committees to be more conservative. Current guidelines (AHA/ACC) suggest that for primary prevention in those >75, it is 'reasonable' (Class IIb) to start a statin after a clinician-patient discussion of potential benefits and risks (like polypharmacy and limited life expectancy), rather than a strong Class I recommendation, due to the nuanced results seen in trials like PROSPER.

Clinical Landscape

Noteworthy Related Trials

2002

Heart Protection Study

n = 20,536 · Lancet

Tested

Simvastatin 40mg daily

Population

Adults at high risk of coronary disease

Comparator

Placebo

Endpoint

All-cause mortality

Key result: Simvastatin significantly reduced the risk of death, nonfatal MI, and stroke across a broad range of high-risk patients.

2003

ASCOT-LLA

n = 10,305 · Lancet

Tested

Atorvastatin 10mg daily

Population

Hypertensive patients with average or lower-than-average cholesterol

Comparator

Placebo

Endpoint

Nonfatal MI and fatal CHD

Key result: Atorvastatin significantly reduced the primary endpoint in patients with hypertension and multiple CV risk factors.

2008

JUPITER Trial

n = 17,802 · NEJM

Tested

Rosuvastatin 20mg daily

Population

Patients with elevated high-sensitivity C-reactive protein and low LDL cholesterol

Comparator

Placebo

Endpoint

Composite of MI, stroke, arterial revascularization, hospitalization for unstable angina, or death

Key result: Rosuvastatin reduced the rate of major cardiovascular events in individuals who appeared healthy but had elevated inflammatory markers.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis