Preventive cardiology for the aging population: how can we better design clinical trials of statins?

ABSTRACT

Introduction

Older adults form a fast-increasing proportion of the world population. However, gains in increasing quantity of life have not been accompanied by similar gains in quality of life. Older people frequently experience frailty, memory problems, and chronic diseases including cardiovascular disease (CVD) and neurodegenerative diseases. Recent trials have demonstrated the efficacy of anti-hypertensive therapy in older populations but failed to show benefits for aspirin.

Area covered

Statins clearly reduce CVD events in middle-aged populations. There seems to be evidence that the effect is similar in primary prevention older populations based on meta-analyses mainly from sub-groups in large trials, but this becomes less clear with increasing age. However, given differences in drug metabolism and possibly efficacy, competing co-morbidities, their effects on mortality, disability, and dementia in this age group remain to be determined.

Expert opinion

Two large trials are now underway to clarify the role of statin therapy in people aged over 70 years using endpoints of mortality, disability, and neurocognitive endpoints as well as standard cardiovascular disease outcomes. They may provide also provide more evidence on how to approach the over 80 year age group.

KEYWORDS:

• Statin
• cardiovascular disease
• disability
• dementia
• ploy-pharmacy
• elderly

1. Background

All novel therapies targeting lipids in cardiovascular disease (CVD) undergo a standard series of trials within their development program. These typically include studies in clinical populations with hyperlipidemia comprising primary and secondary prevention, diabetes and if indicated specific genetic disorders such as familial hypercholesterolemia. More recently studies have included selection of patients with statin intolerance. To generate CVD outcomes, evidence populations from these categories are recruited for the endpoint studies.

The elderly is a fast-increasing proportion of the world population. In the UK, 10 million people are aged over 65 years old and this will double to 19 million by 2050 [1]. However, gains in increasing quantity of life have not been accompanied by similar gains in quality of life. Older people frequently experience frailty, memory problems, and chronic diseases including CVD and neurodegenerative disease [2]. Health care costs in the elderly are associated with co-morbidities and impairments and not simply due to age alone so a reduced age-specific incidence of cardiac failure, stroke, and vascular dementia may delay the need for residential care and thus deliver substantial benefits [3].

Atherosclerosis is primarily a disease of non-modifiable risk factors such as aging or gender with the population attributable risk due to these risk factors being about 60–70% [4] and modifiable risk factors may contribute less in older populations [5]. As a consequence, all risk calculation systems suggest treatment over the age of 60 years in men and 70 years in women irrespective of other CVD risk factors. These recommendations are based on the CVD outcomes trials conducted in predominantly male, middle-aged populations that have been performed with these drugs. Preventive healthcare has reduced the burden of chronic illness at younger ages but there is limited evidence in those aged >70 years [2]. The findings both for efficacy and safety are then extrapolated to older patients based on sub-group analyses. More recently studies have aimed to increase the underlying risk profile of the recruited population often using age greater than 65 years as one of the criteria.

While the case for lipid-lowering therapy in patients with established CVD or at very high risk of developing CVD e.g. concurrent diabetes is well established and can be justified on reduced interventional procedures alone, the situation is not as clear in primary prevention in older patients. Though many older people do have atherosclerotic disease, it is not necessarily their cause of death.

2. Epidemiological studies

Statins are the most commonly used cholesterol lowering medications. Epidemiological studies suggest a benefit of statins in reducing chronic disease and dementia in the elderly [5]. Epidemiological studies suggest that statins may improve rates of motor decline [6], attenuate progression to Alzheimer’s disease [7], or Parkinson’s disease [8]. A meta-analysis of cohort studies (n = 90) found that statin therapy was associated with a 30% (95% confidence interval 14–41%) reduction in the incidence of dementia comprising a 39 (25–50%) fall in Alzheimer’s disease, a 27 (12–39)% fall in other dementias and a 16% (5–26%) reduction in Parkinson’s disease but all with high levels of heterogeneity [9].

There is epidemiological data to suggest that cholesterol levels vary with age and peak at age 40–49 years in men and 50–59 years in women [10] and decline significantly in the last 2 years prior to death [11] which may reflect co-morbidities such as cancer which may not respond to statin therapy [12]. Elderly populations have many risk factors for increased adverse events with statins including a greater predisposition to diabetes, reduced metabolism, drug interactions, low muscle mass, prior muscle pains, and impaired renal and thyroid function [13–15]. However, there is no reason to suppose the process of atherosclerosis and its complications are different in older individuals. A cohort study in Denmark showed little difference in efficacy between older (n = 16035) and younger (n = 49155) populations in a registry study of data from with a median 1.70 mmol/l reduction in LDL-C. Reductions in CVD risk were similar (p = 0.79) at 23 (17–29)% per 1 mmol/LDL-C and 24 (20–29)% in younger individuals censoring at age 70 years [16]. Conversely, a Danish registry study (n = 10388) showed greater reductions in LDL-C than middle-aged populations to standard doses of statins (e.g. 44% vs 40% for atorvastatin 20 mg) [17]. A propensity matched cohort of 5619 Caucasian patients without known prior CVD presenting with new CVD and aged >65 years showed a predictably lower rate of myocardial infarction in statin users with a 15% absolute risk reduction (AR) and 45 (34–55)% relative risk (RR) reduction in those aged 65–75 years and 13% AR reduction with 42 (38–54)% RR reduction in those aged >76 years independent of hypercholesterolemia. The reduction in 30 day CVD mortality was similar in both age groups (ARR 10% RR 61 (32–77)% vs ARR 4%; RR 63 (18–83)%; p = 0.46 for older and younger age groups) but a possible difference existed with the cohort of younger women [18]. Similar data has been reported in a propensity matched cohort of statin commencement in 3670 older patients in a population of 81,729 individuals with and without diabetes [19]. However, another cohort study in Spain of 46,864 participants (mean age 77 years; 63% women; median follow-up 5.6 years) statin use in 75–84 year olds without diabetes was associated with a 6 (−4 to 14)% reduction in CVD events and 2 (−5 to 9)% reduction in all-cause mortality, while in those aged >85 years the effects were 8 (−6 to 18)% and 3 (−5 to 10)%, respectively. Greater effects were seen in participants with diabetes with a 34 (11–35)% reduction in CVD events and 16 (6–25)% in mortality for 75–84 year olds and 18 (−26 to 47)% and −5(−28 to 14)% in those aged >85 years, respectively. A continuous scale analysis using splines corroborated the lack of beneficial statins effect for CVD events and mortality in participants without diabetes older than 74 years [20].

UK primary care databases indicate that statin prescribing in the >80 year age group (n = 212566) has increased from 10% in 2001–5 to 49% in 2011–15 including 12% in centenarians as a result of continuation of scripts. Statins were also more frequently de-prescribed at 6.5%/year in primary prevention populations than in secondary prevention (5.2%/year; p < 0.001). The role of frailty and co-morbidities contribute to this practice. However, frailty and its components, including physical frailty, exhaustion, low physical activity, slow gait speed and weakness, are known CVD risk factors working through a variety of mechanisms [21,22]. Thus, clinical practice tends toward de-prescription in elderly primary prevention populations despite suggestions of continuing CVD benefit [23]. Thus, the balance of risk and benefit for statins remains unclear in this age group [24].

Older people tend to accumulate chronic morbidities with age and thus polypharmacy and drug interactions become more relevant in this population compared to younger age groups [25]. Statin intolerance is a common problem with some surveys suggesting frequencies of up to 27% [15]. A systematic review of 176 studies (112 randomized studies; 64 cohort studies) comprising 4,000,000 individuals suggested a prevalence of 5.9–7.0% according to different definitions. A regression analysis from this study identified age (odds ratio (OR) 1.33, p = 0.04), female gender (OR 1.47, p = 0.007), non-Caucasian ethnicity (p < 0.05), obesity (OR 1.30, p = 0.02), diabetes mellitus (OR 1.26, p = 0.02), hypothyroidism (OR 1.37, p = 0.01), and chronic liver or renal failure (p < 0.05) as well co-prescription of anti-arrhythmic drugs, calcium channel blockers (dihydropyridine class), and statin dose as significant factors [26].

3. Study designs for trials in the elderly

Randomised controlled trials have begun to be done in older populations. For many years, it was thought that higher blood pressure in the elderly simply represented an aging process and that blood pressure lowering would be of limited efficacy. This was disproved in the Hypertension in the Very Elderly (HYVET) study which showed that reducing blood pressure to 150/90 mmHg (a higher target level than that accepted for the middle aged of 140/90 mmHg) in a population aged >75 years reduced not only CVD events but also incidence of hypertensive heart failure leading to a greater overall benefit of treatment [27]. This trial used a classical CVD outcomes endpoint of fatal and non-fatal CVD events. This endpoint is limited to CVD, but wider considerations apply to elderly populations. For many years, it was thought that aspirin therapy should be part of the mandatory CVD prevention package and be formulated within polypills. However, doubts have always existed about the cost–benefit ratio in older populations given the propensity of aspirin to be associated with bleeding episodes or anemia, tinnitus, exacerbation of asthma and drug interaction given the common use of other non-steroidal anti-inflammatory medications for the management of comorbidities such as osteoarthritis.

A more extensive trial design incorporating death, disability and neurological outcomes was used in the Aspirin in Reducing Events in the Elderly (ASPREE) trial [28–30]. This trial recruited 19,114 older adults (10782 females (56%); median (IQR) age, 74 (72–78) years) without overt CVD or diabetes. In ASPREE aspirin therapy for a median of 4.7 years had no effect on rate of CVD events (10.7 vs. 11.3 events per 1000 person-years; Hazard ratio (HR) 0.95 (0.83–1.08)) but did increase the rate of major hemorrhage (8.6 vs. 6.2 events per 1000 person-years; HR 1.38 (1.18–1.62); p < 0.001) [28]. No effect was seen on the incidence of ischemic stroke (HR 0.89 (0.71–1.11)) but there was an increase in intracranial bleeds (108 (1.1%) vs.79 (0.8%); HR 1.38 (1.03–1.84)) [31]. A non-significant increase was seen in hemorrhagic stroke (49 (0.5%) vs. 37 (0.4%); HR 1.33 (0.87–2.04)). No effect was seen on dementia or cognitive decline and mortality increased for unclear reasons [32].

4. Lipids in cardiovascular prevention in the elderly

The CVD prevention package (including co-formulations such as the polypill) had been thought to mandate the use of aspirin, an anti-hypertensive and lipid-lowering therapy in all individuals at increased risk. Studies such as HYVET and ASPREE have confirmed the utility of anti-hypertensives but not aspirin in primary prevention. However, in people without CVD (primary prevention), the benefits of statins are not clear for those aged 70 years or older. Some trials of statins (PROspective Study of Pravastatin in the Elderly at Risk (PROSPER)), The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT), Heart Outcomes Prevention Evaluation-3 (HOPE-3), that have included older people without CVD aged more than 65–70 years, have reported no benefits whereas others have reported large benefits. Pravastatin in the PROSPER trial in 3,239 people over 3.2 years, which matched its initial screened population well [33] but included 44% with CVD, reduced composite CVD events by 15 (3–26)% (p = 0·01). Coronary heart disease death and non-fatal myocardial infarction but did not significantly reduce major fatal CVD events ((15(−7 to 33)%; p = 0.16)) except in post-trial follow-up [34]. Similarly, in the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial-Lipid Lowering Trial (ALLHAT-LLT) in 2867 people treated for 4.5 years, pravastatin did not significantly reduce major CVD events, had high rates of confounding drop-in and drop-out to statins and raised concern of potential harm (increased mortality) in people aged >75 years [35,36]. Rosuvastatin in the HOPE-3 Trial in 3806 people aged greater than 70 years for 5 years did not significantly reduce CVD events showing a 17 (−7 to 34)% reduction in the combined CVD end point (p = 0.16), and 9(−13 to 27)% reduction in all-cause mortality [37]. However, in the Justification for the Use of Statins in Primary Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER) trial, of 5695 people over 2 years, rosuvastatin reduced CVD events by 39(14–57)% (p = 0.004) and a non-significant 20% reduction in all-cause mortality consistent with effects in younger age groups [37]. A pooled analysis from JUPITER and HOPE-3 suggested that statins reduce CVD events in people >70 years by 24 (11–31)% but provided no data on higher age subgroups [37]. In ASPREE 5,629 patients took statins at baseline. Over a median follow-up period of 4.7 years, baseline statin use was not associated with disability-free survival or with the risk for all-cause mortality or dementia (HR 0.92 (0.83–1.03); p = 0.17). Statin therapy was associated with lower risks for physical disability (HR 0.75 (0.58–0.96); p = 0.02) and CVD outcomes (HR 0.68 (0.57–0.82); p < 0.001). More recently, a subgroup analysis for 574 patients (15%) aged >75 years in the Randomized Comparison of Efficacy and Safety of Lipid-lowerING With Statin Monotherapy Versus Statin/Ezetimibe Combination for High-risk Cardiovascular Diseases (RACING) trial showed no difference in the rate of CVD events but moderate-intensity statin with ezetimibe combination therapy compared with high-dose monotherapy was associated with a suggestion of lower rates of intolerance-related drug discontinuation or dose reduction among patients aged >75 years (2.3 vs 7.2%; p = 0.01) but not significantly more compared to those <75 years (5.2 vs 8.4%; p < 0.001) (P_interaction = 0.16) [38].

There is little specific clinical trial evidence outside statins in older patients. The Ezetimibe Lipid-Lowering Trial on Prevention of Atherosclerotic Cardiovascular Disease in 75 or Older (EWTOPIA 75) recruited 3796 patients aged >75 yrs (age 85 ± 5 yrs), randomized them to ezetimibe therapy and followed them up for a median 4.1 years using a blinded outcome design. The trial included 74% women, 89% with hypertension and 25% with diabetes. Ezetimibe reduced LDL-C by 26% (1.10 mmol/L) and the primary outcome of CVD events and interventions by 34 (14–50)% (p = 0.002) with number needed to treat of 38. Secondary outcomes of coronary events were reduced by 40 (2–63)% (p = 0.04) and coronary revascularization by 62(21–82)% (p = 0.007) with no difference in the incidence of stroke, all-cause mortality, or adverse events [39].

A meta-analysis of 21,492 patients aged >75 years from subgroups of 29 trials comprising 11,750 (55%) from statin trials, 6209 (29%) from ezetimibe trials, and 3533 (16%) from PCSK9 inhibitor trials with follow-up 2.2 to 6.0 years has been done [40]. Major CVD events (n = 3519) were reduced by 26% per 1 mmol LDL-C (RR 0.74(0.61–0.89); p = 0.002) similar to younger patients (RR 0.85 (0.78–0.92); p_interaction = 0.37). Effects were similar for statin (0.82 (0.73–0.91)) and non-statin treatments (RR 0·67(0·47–0·95); p_interaction = 0·64) and for each component of the outcome including CVD death (RR 0.85 (0.74–0.98)), myocardial infarction (RR 0.80 (0.71–0.90)), stroke (RR 0.73 (0.61–0.87)), and coronary revascularisation (RR 0.80(0.66–0.96)). However, analyses of Cholesterol Treatment Trialists’ database suggest that these benefits are dependent on the age cutoffs chosen for these meta-analyses as the benefits of statin therapy seem to reduce with increasing age [41]. Meta-analyses in the over-75 yrs age group also can give different results depending on whether classical frequentist or Bayesian analyze are used [42].

4.1. CVD guidelines and the elderly

Most guidelines including those from the National Institute for Care and Health Excellence (NICE) have not been able to make strong recommendations for older people and any recommendations have generally been extrapolated from trials in younger people (<65 years). As guidelines that have moved from target-based to absolute CVD-risk-based treatment strategies, these controversially risk medicalization of ‘healthy populations’ as all elderly individuals (>70 years) have a CVD risk >10% [43]. These guidelines suggest that statin therapy should be considered in older patients in higher risk categories or multiple-risk factors. The Cholesterol Treatment Trialists meta-analysis calculated an absolute risk reduction of 0.6%/year per 1 mmol/L LDL-C levels in patients aged >75 years, that would translate into a number needed to treat (NNT) of 167. However, as absolute risk is greater in older patients due to the age contribution to CVD risk, the absolute effect of 1 mmol/L reduction is potentially 10-fold greater in older as opposed younger patients [44]. There is growing debate on whether otherwise healthy older people should be prescribed drugs to prolong life and reduce disability. Given the uncertainty, there is a need for a randomized controlled trial to answer the questions of relevance to an aging population. This was highlighted in the Research recommendations accompanying the UK NICE guidelines for Lipid Modification and Cardiovascular Risk Assessment (CG181) [13] and by the US National Institute for Aging.

5. Clinical trials in progress

A couple of studies are now underway to investigate the extent of the role for statin therapy in older primary prevention populations.

The Statins in Reducing Events in the Elderly (STAREE) trial is a double-blind randomized placebo-controlled study involving community-dwelling men and women >70 years which has recruited 12,000 primary prevention participants who have been randomized to atorvastatin 20 mg titrated to 40 mg with down titration allowed if necessary [45]. Follow-up is expected to be an average of 5 years from randomization. The co-primary endpoints are (i) death or development of dementia or development of physical disability, and (ii) major fatal or non-fatal CVD event (non-fatal myocardial infarction, non-fatal stroke, CVD death). Additional data is being collected on hospitalization events; new diagnosis of diabetes; fatal and non-fatal cancer (excluding non-melanoma skin cancer); the need for permanent residential care; all-cause dementia, frailty/Disability, cognitive decline and quality of life.

The Pragmatic evaluation of events and benefits of lipid lowering in older adults (PREVENTABLE) trial is comparing high-intensity statin therapy (atorvastatin 40 mg) with placebo in 20,000 primary prevention patients aged over 75 years [46]. Exclusion criteria include statin use in the prior year or for >5 years and inability to take a statin. The primary outcome measure is survival free of new dementia or persisting disability while co-secondary outcomes include CVD events (including heart failure and interventions) and a composite of mild cognitive impairment or dementia.

6. Conclusions

Statins clearly reduce CVD events in middle-aged populations. There seems to be evidence that the effect may be similar in older populations based on meta-analyses. However, given differences in drug metabolism and possibly efficacy, competing co-morbidities, their effects on mortality, disability and dementia in this age group remain to be determined. Two large trials are now underway to clarify the role of statin therapy in people aged over 70 years.

7. Expert opinion

There are competing schools of thought about special populations. In some guidelines or in risk calculators, ethnicity has been used to modify biomarker (e.g. creatinine clearance) or CVD event risk but these modifications have secondary consequences in terms of access and equality so non-ethnicity-based systems are being increasingly preferred. Similarly, age has long been used as a risk modifier, but its limitations are well known. Children are considered a special case and not just as ‘young adults’ with separate drug licensing requirements due to differences in drug metabolism, potential effects on growth and maturation and the use of centile charts or relative risk as a better guide to future risk. There is separate guidance for use of statins in children both generally and in familial hypercholesterolemia [47]. At the other extreme the elderly have been considered as ‘older adults’. However, older adults differ from middle-aged populations in a greater incidence of chronic disease and hence polypharmacy, potential differences in drug metabolism and increased prevalence of chronic disability manifesting as frailty and neurodegenerative disease. This has been summarized in the geriatrics 5 M strategy of mind (cognition, memory), mobility (balance, falls, and function), medications (polypharmacy, adverse effects, deprescribing), multi-complexity (multi-morbidity, biopsychosocial context), and matters most (individual goals and preferences) [48]. Trials of aspirin in CVD prevention show no benefit and increased risks of anemia in older adults raising questions about the composition of any CVD poly-pill for this age group. At one extreme this combination could be as brief as an angiotensin receptor antagonist and a statin or extend to aspirin, 2 or more anti-hypertensive drugs, 1or more cholesterol-lowering drugs some at very low doses. Some guidelines e.g. hypertension where clinical trials prove the benefit of intervention (HYVET) recognize that frailty and vascular aging (increased arterial stiffness) mean that considerations such as risks of postural hypotension are relevant to prescribing. However, the role of frailty is complex as it is a CVD risk factor or a coincidental marker of underlying disease [49] and a potential limit to treatment. To date almost all the evidence for lipid intervention from randomized controlled trials is based on sub-populations and meta-analyses. Statin therapy shows clear benefits in reducing CVD events especially repeat intervention in secondary prevention populations and even in approximate CVD risk equivalents such as type 2 diabetes but there is little evidence in primary prevention. Current CVD risk calculators have been optimized for sensitivity as intervention thresholds have been reduced but this has the consequence that all older adults qualify for statin therapy as age is the primary determinant of risk and implying over-treatment and hence excess avoidable adverse events. The extent of coronary atheroma burden varies with age and not all older individuals have imaging-identifiable disease [50]. This has led to suggestions that the relative risk approach and individualized decision making may be more useful in this age group [51]. Furthermore, the contribution of modifiable risk CVD factors to overall risk may fall with age reducing the benefits of treatment [5,28]. Therefore, questions remain about the utility of statin therapy in primary prevention of CVD in older adults and their relative benefits on CVD death and morbidity. Whether statins have effects on other chronic diseases such as cancer, neurodegenerative disease (e.g. Parkinson’s and dementia) is also unclear as the epidemiological evidence is confusing. Specific studies such as STAREE and PREVENTABLE are designed to answer these questions and extend the evidence base into this age group. However, these studies have their own limitations being designed for relatively low-risk populations by setting the recruitment from age 70 years (STAREE) or 75yeras (PREVENTABLE) and excluding co-morbidities such as diabetes, prior chronic disease such as heart failure or neurodegenerative disease or those likely to result in a high probability of death from established disease in the trial. One of the consequences of this selection, as seen in ASPREE, is to identify a patient group with a lower incidence of CVD and higher one of cancer [30]. These studies will provide clear data on comparison with placebo for CVD, disability-based and mortality outcomes. Suggestions have been made to conduct trials in the over 80 years age group with minimal exclusions and electronic health record-based data gathering but would face difficulties from allocation to placebo of those at high calculated CVD risk or confounding by other causes. They may be possible if baseline lipid-lowering therapy with statins is allowed in both arms and the comparison is with intensified therapy using ezetimibe or more effectively pro-protein convertase subtilisin kexin-9 (PCSK9) inhibitors where long-acting injections (e.g. inclisiran) or newer oral agents (MK-0616) may be preferred.

Article highlights

• Most data on older patients are derived from sub-groups of younger individuals recruited to clinical trials.

• Older patients are more likely to have other co-morbidities and be subject to polypharmacy and may have different drug metabolism compared with younger individuals.

• Statin therapy has been associated with benefits of non-cardiovascular chronic diseases in cohort studies in older patients.

• Standard cardiovascular event endpoints are not the only appropriate measure of effectiveness in this age group.

• Two randomized controlled clinical trials of statin therapy in primary prevention are underway in this population.

Declaration of interest

A Wierzbicki is an investigator in the STAREE trial. The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

A peer reviewer on this manuscript has collaborated with several companies marketing CVD drugs. Peer reviewers on this manuscript have no other relevant financial relationships or otherwise to disclose.

Additional information

Funding

This paper was not funded.