Michael Rocco, MD

Statin therapy has been shown to substantially reduce adverse events associated with low-density-lipoprotein cholesterol (LDL-C) and cardiovascular disease (CVD). Statins alone are often not adequate to achieve treatment goals, and residual CVD risk remains high. Combination therapies of statins with ezetimibe and resins to further lower LDL-C, fibrates and omega 3 fatty acids to lower triglycerides, and niacin to lower both and raise high-density-liproprotein cholesterol are available, but additional risk reduction has not been consistently demonstrated in clinical trials.

The link between atherogenic lipoproteins and CVD is strong, and the need to develop therapies in addition to statins to substantially and safely reduce LDL-C is a priority. The association of reduced proprotein convertase subtilisin/kexin type 9 (PCSK9) activity with reduced LDL-C and CVD events has led to the rapid development and approval of mono­clonal antibody therapies to inhibit PCSK9.

In this review, we discuss trials of these therapies that have shown durable reductions in LDL-C of more than 50%, with acceptable tolerability. Now that PCSK9 inhibitors are approved by the US Food and Drug Administration (FDA), extended data are needed as to long-term tolerability, safety, and efficacy of these agents and, most importantly, demonstration of additional reduction in CVD events.

A CASE FOR ADDITIONAL THERAPIES

CVD is the leading cause of morbidity and death in the United States, responsible for one in four deaths. Hyperlipidemia and, specifically, elevated LDL-C have been found to be important drivers of atherosclerosis and, in turn, adverse cardiovascular (CV) events. Likewise, numerous observational and clinical trials have shown that reducing LDL-C, particularly with statins, decreases CVD events.^1–4 More aggressive lowering with higher doses or more intensive statin therapy further reduces rates of adverse outcomes.^3,4 In addition, the pleiotropic effects of statins imply that not all of their benefits are derived from LDL-C lowering alone.⁵ Consequently, it is now standard practice to use statins at the highest tolerable dose to reach target LDL-C levels and prevent CV events in high-risk patients with CVD or multi­ple coronary artery disease risk factors, regardless of the LDL-C levels.^6,7

The American College of Cardiology (ACC) and the American Heart Association released cholesterol guidelines in 2013 that recommend a risk-based approach for statin therapy rather than targeting specific LDL-C levels.⁶ Although this evidence-based approach may better conform to clinical trials, the debate that lower LDL-C targets will further prevent CVD continues.

Indeed, it appears that lower is better, as demonstrated by the IMPROVE-IT trial.⁸ Although the control group receiving simvastatin monotherapy had low LDL-C levels (mean, 69.9 mg/dL; 1.8 mmol/L), the experimental group receiving simvastatin plus ezetimibe achieved even lower levels (mean, 53.2 mg/dL;1.4 mmol/L) after 1 year of therapy and had a significantly lower composite primary end point of CV death, major coronary event, or nonfatal stroke at 7 years (34.7% for simvastatin monotherapy vs 32.7% for combined therapy).⁹ Furthermore, the event-rate reduction with the addition of ezetimibe was the same as the average predicted by the Cholesterol Treatment Trialists’ meta-analysis: an LDL-C reduction of 1 mmol/L (38.6 mg/dL) yields a 23% risk reduction in major coronary events over 5 years.¹⁰ Although only a modest absolute reduction in outcomes, it supports the notion that further reduction of LDL-C levels by more potent therapies may offer greater benefit.

There is strong evidence that statin therapy reduces the risk of developing CVD in patients with or without a previous atherosclerotic event; however, residual CVD risk remains even for those on therapy. A contributing factor to this residual risk is that many statin-treated patients have insufficient response or intolerance and do not achieve adequate LDL-C reductions.

There are three clinically important patient populations who are inadequately managed with current therapies and remain at high risk of subsequent CV events; these are patients who would benefit from additional therapies.

1. Patients with familial hypercholesterolemia (FH). This is the most common genetic disorder in the world, yet it is frequently undiagnosed and untreated. Due to high baseline cholesterol levels, achieving LDL-C treatment goals is challenging.

• The prevalence may be closer to 1:200 to 1:250 rather than the often quoted 1:500.¹¹
• Fewer than 12% of patients with heterzygous FH achieve the LDL-C goal of < 100 mg/dL with maximal statin treatment alone or with a second agent.¹²

2. Patients with hyperlipidemia not due to FH who are at elevated CV risk and undertreated. In US and European surveys, between 50% and 60% of patients receiving statins with or without other therapies failed to reach LDL-C reduction goals.¹³

• Variation in response to statin treatment between individuals may be considerable.
• Poor adherence to statin therapy is common.

3. Patients with side effects to statins, particularly muscle symptoms that prevent statin use or substantially limit the dose.

• Although the incidence of myopathy is low (< 0.1%) and rhabdomyolysis is even less common, observational studies suggest that 10% to 20% of patients may limit statin use due to muscle-associated complaints including muscle aching, cramps, or weakness.¹⁴
• Side effects may be dose-dependent, limiting the use of the high-intensity statin doses that are frequently necessary to achieve LDL-C goals.

Consequently, there is great interest in developing therapies beyond statins that may further reduce CV events. However, treatments other than ezetimibe for further management of hyperlipidemia and risk reduction have failed to demonstrate consistent benefit when added to statin therapy.^15–19 The largest studies were with niacin and fibrates. Unfortunately, most trials demonstrated no overall outcomes benefit or only benefits in subgroup analyses, leaving the door open to other pharmacologic interventions.

Studies with the cholesterol ester transfer protein (CETP) inhibitor torcetrapib, in combination with statin therapy, actually demonstrated an overall increase in all-cause mortality in the treatment group.²⁰ Two large outcome trials of the CETP inhibitors dalcetrapib and evacetrapib were stopped after interim analysis predicted no benefit. Although drugs such as lomitapide (a microsomal triglyceride transfer protein inhibitor) and mipomersen (an antisense oligonucleotide inhibitor of ApoB-100 synthesis) can lower LDL-C by reducing ApoB synthesis,²¹ they are approved only in the small population of individuals with homozygous FH and liver toxicity and side effects are a concern.

Accordingly, current cholesterol management guidelines continue to offer LDL-C as the main target of lipid-modifying therapy, with statins as the primary treatment choice. The desire to build on statin therapy to prevent further progression of atherosclerosis and clinical CVD has encouraged continued focus on strategies to lower LDL-C to even greater extents.

Fortunately for practitioners, for the first time since lovastatin was approved in 1987, there is a new therapy approved by the FDA that significantly lowers LDL-C and, potentially, improves CV outcomes—the proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors. This review will focus on the PCSK9 inhibitors, a novel therapeutic class that reduces LDL-C through increased hepatic clearance. These drugs are rapidly emerging as an ideal adjunctive therapy to statins for patients at the highest risk and as a highly efficacious alternate therapy in patients intolerant of statins.

PCSK9 INHIBITORS: DISCOVERY, MECHANISM, AND THERAPEUTIC INTERVENTIONS

Two PCSK9 inhibitors have received FDA approval: alirocumab (Praluent) and evolocumab (Repatha). Among new molecular entities for clinical use, PCSK9 inhibitor therapies had one of the shortest durations from discovery to development and approval.

Mutations in the PCSK9 gene associated with autosomal dominant hypercholesterolemia were first identified in 2003 in a French family.²² The PCSK9 protein is now known to be a secreted enzymatic serine protease that is primarily synthesized in the liver and binds to the LDL receptor (LDL-R)/LDL-C complex on the surface of hepatocytes, marking the receptor for lysosomal degradation rather than recycling to the cell surface. Thus, it reduces the quantity of LDL-R that is available to remove LDL-C from circulation.²³ As a result, higher levels of PCSK9 are associated with higher levels of plasma LDL-C.

The clinical importance of PCSK9 in regulating LDL-C is supported by observed mutations and polymorphisms. Gain-of-function mutations that increase the activity of PCSK9 have been shown to be associated with elevated LDL-C, premature CVD, and myocardial infarction (MI).²⁴ Conversely, loss-of-function mutations (heterozygotes found in 1% to 3% of the population) result in decreased activity of PCSK9, lower LDL-C, and lower incidence of CVD (Table 1).^25–29 These observations, combined with data showing that homozygote loss-of-function individuals with very low LDL-C were generally very healthy, sparked interest in developing inhibition of PCSK9 activity as a therapeutic strategy for hyperlipidemia.

Multiple pharmacologic developments are aimed at inhibiting PCSK9, with many compounds in clinical trials. The approaches include gene silencing with loss-of-function mutations, synthetic peptides, oral small molecules, and monoclonal antibodies. Gene silencing was first observed in 2007 when administration of antisense oligonucleotides targeted to selectively inhibit PCSK9 mRNA was found to up-regulate LDL-R, thereby decreasing serum levels of LDL-C.³⁰

The first study to establish the role of synthetic peptides in PCSK9 inhibition was performed in 2008. In this study, the epidermal growth factor-like A synthetic peptide blocked the interaction between PCSK9 and LDL-R, thereby decreasing the degradation of LDL-R and preserving LDL uptake.³¹ Although studies are limited, synthetic peptides remain an area of great interest given their promising effects on lipid metabolism. Recently, a synthetic PCSK9-binding adnectin derived from the human fibronectin known as BMS-962476, had favorable results in a phase 1 clinical trial. An RNA interference molecule, subcutaneous ALN-PSC, inhibits PCSK9 gene expression by causing destruction of messenger RNA, thus inhibiting PCSK9 synthesis (Table 2).³²

PCSK9 INHIBITORS: CLINICAL TRIALS

Subcutaneously administered monoclonal antibodies targeting PCSK9 currently are the only PCSK9 inhibitors FDA-approved for clinical use. The first study to demonstrate efficacy in enhancing uptake of serum LDL-C was performed in 2009.³³ Multiple phase 1 and 2 studies soon followed, demonstrating acceptable safety and 50% to 70% reductions in LDL-C at upper-dose titrations.³⁴ Additionally, there were significant reductions in total cholesterol, ApoB, triglycerides, and lipoprotein(a).

These early developments paved the way for larger phase 3 trials (Table 3).^35–48 The PCSK9 inhibitors evolocumab and alirocumab have been shown in multiple phase 3 clinical trials to achieve a consistent dose-dependent 50% to 60% reduction in LDL-C across a broad range of CVD risk, pretreatment LDL-C levels, and background therapy: monotherapy (MENDEL-2, ODYSSEY COMBO I),^35,44 added to statin therapy (LAPLACE-2, ODYSSEY CHOICE I),^38,46 and in individuals with heterozygous FH (RUTHERFORD-2, ODYSSEY-FH).^37,42 Trials with bococizumab are under way.

The GAUSS-2 clinical trial (Goal Achievement after Utilizing an Anti-PCSK9 Antibody in Statin Intolerant Subjects-2) demonstrated similar efficacy in reducing LDL-C in patients with clinically assessed statin intolerance due to muscle-related adverse symptoms.³⁹ In GAUSS-3, patients were first identified as being statin intolerant secondary to muscle-associated symptoms based on a randomized, crossover trial of atorvastatin vs placebo.⁴⁰ The 43% of participants who experienced intolerable muscle-related symptoms on the statin but not on placebo were then randomized to evolocumab vs ezetimibe. Results showed significant reduction in LDL-C in the evolocumab group (52.8%) compared with the ezetimibe group (16.7%). Additionally, among patients with muscle symptoms on statin therapy, PCSK9 therapy was discontinued for muscle symptoms in only 0.7% of evolocumab recipients and 6.8% of ezetimibe recipients.

Overall, the PCSK9 inhibitors are generally well tolerated with injection site reactions being the most common side effect. A meta-analysis published in 2015 of 25 trials including more than 12,000 patients treated with evolocumab and alirocumab reported no significant difference in adverse events or safety outcomes vs placebo or ezetimibe.⁴⁹ Antidrug binding or neutralizing antibody production to these agents, thus far, has not been shown to be an issue. Additional analyses have not indicated an adverse effect on gonadal hormone levels or increased incidence of new-onset diabetes.

Two studies published in 2015 offer insight into longer term durability and safety as well as potential CVD outcome benefit (Table 4)^50,51:

• OSLER-1 and 2: Open-Label Study of Long-Term Evaluation against LDL-Cholesterol (OSLER) trials—evolocumab trial;⁵⁰
• ODYSSEY long term: Long-Term Safety and Tolerability of Alirocumab in High Cardiovascular Risk Patients with Hypercholesterolemia Not Adequately Controlled with Their Lipid Modifying Therapy—alirocumab trial.⁵¹

The OSLER trials reported durable LDL-C reductions of 61% and the ODYSSEY trial reported a LDL-C reduction of 62%.^50,51 In both studies, the overall occurrence of adverse events was similar to placebo, but both reported a higher rate of neurocognitive effects in the active treatment groups (evolocumab 0.9% vs 0.3% for standard therapy; alirocumab 1.2% vs 0.5% for placebo). It must be noted that although the absolute rate of neurocognitive adverse events is low, it is unclear if these events were related to the drugs themselves or to extreme lowering of LDL-C. Nevertheless, the FDA has raised concerns about neurocognitive events. A sub-study of the ongoing FOURIER trial with evolocumab—EBBINGHAUS—is expected to address this concern.

In addition, analyses of CV events showed that the PCSK9 inhibitors effectively cut the CV rate in half in both studies (Figure 1).^50,51 In the OSLER trials,⁵⁰ evolocumab recipients had 53% reduction in major CV events (0.95% vs 2.18% in the standard therapy group; P = .003). In ODYSSEY,⁵¹ alirocumab recipients had a 48% reduction in major CV events (1.7% vs 3.3% for placebo; P = .02). Furthermore, a 2015 meta-analysis of 24 phase 2 and 3 trials reported a statistically significant 55% reduction in all-cause mortality and 50% reduction in CV mortality with PCSK9 inhibitors.⁵²

For many reasons including short length of follow-up, study design, and small numbers of outcome events, the OSLER and ODYSSEY studies, although enticing, are exploratory and hypothesis-generating only and results need to be interpreted with caution. Nevertheless, they have set the stage for ongoing prospective randomized outcome trials that are studying the CV effects and tolerability of PCSK9 inhibitors over a longer time frame. These include the following trials.

• The Further Cardiovascular Outcomes Research with PCSK9 Inhibition in Subjects With Elevated Risk (FOURIER) is an ongoing trial with the primary end point of CV death, MI, hospitalization for unstable angina, stroke, or coronary revascularization in high-risk patients receiving evolocumab or placebo.⁵³
• The ODYSSEY trial is examining the effect of alirocumab vs placebo on the composite primary endpoint of coronary heart disease death, non-fatal MI, fatal and nonfatal ischemic stroke, and unstable angina requiring hospitalization in patients who have had an acute coronary syndrome event during the previous 4 to 52 weeks.⁵⁴
• The Evaluation of Bococizumab in Reducing the Occurrence of Major Cardiovascular Events in High Risk Subjects (SPIRE) trials are investigating the effect of bococizumab, a third PCSK9 “humanized” monoclonal antibody, vs placebo in reducing death, MI, stroke, or unstable angina in patients at high-risk of CVD who are receiving standard lipid-lowering therapy with LDL-C > 70 mg/dL (1.8 mmol/L) (SPIRE-1) or > 100 mg/dL (2.6 mmol/L) (SPIRE-2).^55,56  

Because these outcome trials are attempting to enroll more than 70,000 patients and are event driven, it is difficult to predict when they will be completed (Table 5).^53–56 However, recent estimates indicate completion of at least one trial by the end of 2016 or early 2017, with interim analyses of others expected at that time. It is hoped that they will answer the all-important question of whether PCSK9 inhibitors are associated with further CV event reduction benefit.

CURRENT FDA INDICATIONS AND GUIDELINES

The two PCSK9 inhibitors approved by the FDA—alirocumab (subcutaneous 75 mg every 2 weeks up titrated to 150 mg) and evolocumab (subcutaneous 140 mg every 2 weeks or 420 mg every 4 weeks)—are both indicated for use with statins in patients with heterozygous FH or known atherosclerotic CVD who require further reduction in LDL-C levels despite lifestyle interventions and use of maximally tolerated statins. Evolocumab has also been approved for use in patients with homozygous FH.

Although PCSK9 inhibitors are not specifically approved for patients unable to tolerate statins, the results of GAUSS-3, which documented that statin intolerance is a real, definable entity and very responsive to PCSK9 inhibition, makes these drugs promising agents for patients intolerant of statins and, thus, unable to benefit from high-intensity stain therapy.

In April 2016, the ACC released a clinical consensus update to their 2013 cholesterol guidelines, which is their first recommendation specifically addressing the use of non-statin therapies, including the newer PCSK9 inhibitors.⁵⁷ For high-risk patients with clinical atherosclerotic CVD or LDL-C > 190 and failure to achieve at least a 50% reduction in LDL-C on maximally tolerated statin, non-statins may be considered. Ezetimibe, given its safety and tolerability, should be the first additional medication added. Bile acid sequestrants may be used as a second-line therapy if ezetimibe is not tolerated and triglycerides are not elevated. If therapy goals are not met on maximally tolerated statin and ezetimibe, either approved PCSK9 inhibitor can be added or used to replace ezetimibe. The document also specifies that given the lack of long-term safety and efficacy data on the PCSK9 inhibitors, they are not recommended for use in primary prevention patients in the absence of FH.

CONCLUSION

Although statin therapy has been shown to substantially reduce LDL-C and CVD adverse events, there remains a high rate of inadequate goal achievement and residual CVD risk in patients receiving statins. Combination therapies with ezetimibe and resins to further lower LDL-C, fibrates and omega 3 fatty acids to lower triglycerides, and niacin to lower both and raise high-density-liproprotein cholesterol are available, even though additional CV risk reduction is minimal or elusive when these drugs are added to statin therapy.

The link between atherogenic lipoproteins and CVD is strong, and the need to develop therapies in addition to statins to substantially and safely reduce LDL-C remains a priority. The association of reduced PCSK9 activity with reduced LDL-C and CV events has led to rapid development and approval of monoclonal antibody therapies to inhibit PCSK9. In trials, these therapies have shown substantial and durable reductions in LDL-C of more than 50%, with acceptable tolerability. Now that PCSK9 inhibitors are approved by the FDA, extended data about long-term tolerability, safety, and efficacy and, most importantly, demonstration of additional reduction in CVD events are needed. It is hoped that the long-term ongoing trials will provide these data.

For the immediate future, statin therapy will continue to be the cornerstone of lipid and CVD risk management based on their low generic cost, proven CVD risk reduction, and clinicians’ comfort with their use. However, the reliable efficacy of PCSK9 inhibitors and the fact that statin therapy itself increases PCSK9 activity makes the addition of PCSK9 inhibitors to statins an attractive approach in high-risk patients failing to reach LDL-C treatment goals.

Although current indications are limited, there are patients at high CVD risk who would be appropriate candidates for these therapies. These include patients with the following:

• FH with lifetime burden of elevated LDL-C and associated low likelihood of achieving optimal LDL-C control on current available therapies
• Complete or partial statin intolerance with high-intensity statin dosing limited by side effects
• High CV risk who are not at LDL-C goal on current therapies.

Now that the first therapies are available, practitioners can expect newer approaches to tackle PCSK9-mediated LDL-C reduction. Bococizumab is lagging in phase 3 trials, but the SPIRE program is moving forward with special population studies expected to conclude in 2016 and simultaneous long-term outcomes trials. Other PCSK9 inhibitors being investigated include agents with more durable effect requiring less frequent injections, RNA-interference therapies, vaccinations, antisense therapies, and oral formulations.

The PCSK9 inhibitors hold promise as an adjunct to statin therapy. Their eventual clinical role will depend on a balance between substantial LDL-C reductions, long-term safety, tolerability, and reduction in CVD events vs the cost (estimated at $14,000 a year), access from payers, acceptance of injectable therapies, and magnitude of incremental benefit when added to current therapies. Nevertheless, initial clinical trial data are encouraging and these drugs may be an important addition to the therapeutic armamentarium against CVD.

Statin therapy has been shown to substantially reduce adverse events associated with low-density-lipoprotein cholesterol (LDL-C) and cardiovascular disease (CVD). Statins alone are often not adequate to achieve treatment goals, and residual CVD risk remains high. Combination therapies of statins with ezetimibe and resins to further lower LDL-C, fibrates and omega 3 fatty acids to lower triglycerides, and niacin to lower both and raise high-density-liproprotein cholesterol are available, but additional risk reduction has not been consistently demonstrated in clinical trials.

The link between atherogenic lipoproteins and CVD is strong, and the need to develop therapies in addition to statins to substantially and safely reduce LDL-C is a priority. The association of reduced proprotein convertase subtilisin/kexin type 9 (PCSK9) activity with reduced LDL-C and CVD events has led to the rapid development and approval of mono­clonal antibody therapies to inhibit PCSK9.

In this review, we discuss trials of these therapies that have shown durable reductions in LDL-C of more than 50%, with acceptable tolerability. Now that PCSK9 inhibitors are approved by the US Food and Drug Administration (FDA), extended data are needed as to long-term tolerability, safety, and efficacy of these agents and, most importantly, demonstration of additional reduction in CVD events.

A CASE FOR ADDITIONAL THERAPIES

CVD is the leading cause of morbidity and death in the United States, responsible for one in four deaths. Hyperlipidemia and, specifically, elevated LDL-C have been found to be important drivers of atherosclerosis and, in turn, adverse cardiovascular (CV) events. Likewise, numerous observational and clinical trials have shown that reducing LDL-C, particularly with statins, decreases CVD events.^1–4 More aggressive lowering with higher doses or more intensive statin therapy further reduces rates of adverse outcomes.^3,4 In addition, the pleiotropic effects of statins imply that not all of their benefits are derived from LDL-C lowering alone.⁵ Consequently, it is now standard practice to use statins at the highest tolerable dose to reach target LDL-C levels and prevent CV events in high-risk patients with CVD or multi­ple coronary artery disease risk factors, regardless of the LDL-C levels.^6,7

The American College of Cardiology (ACC) and the American Heart Association released cholesterol guidelines in 2013 that recommend a risk-based approach for statin therapy rather than targeting specific LDL-C levels.⁶ Although this evidence-based approach may better conform to clinical trials, the debate that lower LDL-C targets will further prevent CVD continues.

Indeed, it appears that lower is better, as demonstrated by the IMPROVE-IT trial.⁸ Although the control group receiving simvastatin monotherapy had low LDL-C levels (mean, 69.9 mg/dL; 1.8 mmol/L), the experimental group receiving simvastatin plus ezetimibe achieved even lower levels (mean, 53.2 mg/dL;1.4 mmol/L) after 1 year of therapy and had a significantly lower composite primary end point of CV death, major coronary event, or nonfatal stroke at 7 years (34.7% for simvastatin monotherapy vs 32.7% for combined therapy).⁹ Furthermore, the event-rate reduction with the addition of ezetimibe was the same as the average predicted by the Cholesterol Treatment Trialists’ meta-analysis: an LDL-C reduction of 1 mmol/L (38.6 mg/dL) yields a 23% risk reduction in major coronary events over 5 years.¹⁰ Although only a modest absolute reduction in outcomes, it supports the notion that further reduction of LDL-C levels by more potent therapies may offer greater benefit.

There is strong evidence that statin therapy reduces the risk of developing CVD in patients with or without a previous atherosclerotic event; however, residual CVD risk remains even for those on therapy. A contributing factor to this residual risk is that many statin-treated patients have insufficient response or intolerance and do not achieve adequate LDL-C reductions.

There are three clinically important patient populations who are inadequately managed with current therapies and remain at high risk of subsequent CV events; these are patients who would benefit from additional therapies.

1. Patients with familial hypercholesterolemia (FH). This is the most common genetic disorder in the world, yet it is frequently undiagnosed and untreated. Due to high baseline cholesterol levels, achieving LDL-C treatment goals is challenging.

• The prevalence may be closer to 1:200 to 1:250 rather than the often quoted 1:500.¹¹
• Fewer than 12% of patients with heterzygous FH achieve the LDL-C goal of < 100 mg/dL with maximal statin treatment alone or with a second agent.¹²

2. Patients with hyperlipidemia not due to FH who are at elevated CV risk and undertreated. In US and European surveys, between 50% and 60% of patients receiving statins with or without other therapies failed to reach LDL-C reduction goals.¹³

• Variation in response to statin treatment between individuals may be considerable.
• Poor adherence to statin therapy is common.

3. Patients with side effects to statins, particularly muscle symptoms that prevent statin use or substantially limit the dose.

• Although the incidence of myopathy is low (< 0.1%) and rhabdomyolysis is even less common, observational studies suggest that 10% to 20% of patients may limit statin use due to muscle-associated complaints including muscle aching, cramps, or weakness.¹⁴
• Side effects may be dose-dependent, limiting the use of the high-intensity statin doses that are frequently necessary to achieve LDL-C goals.

Consequently, there is great interest in developing therapies beyond statins that may further reduce CV events. However, treatments other than ezetimibe for further management of hyperlipidemia and risk reduction have failed to demonstrate consistent benefit when added to statin therapy.^15–19 The largest studies were with niacin and fibrates. Unfortunately, most trials demonstrated no overall outcomes benefit or only benefits in subgroup analyses, leaving the door open to other pharmacologic interventions.

Studies with the cholesterol ester transfer protein (CETP) inhibitor torcetrapib, in combination with statin therapy, actually demonstrated an overall increase in all-cause mortality in the treatment group.²⁰ Two large outcome trials of the CETP inhibitors dalcetrapib and evacetrapib were stopped after interim analysis predicted no benefit. Although drugs such as lomitapide (a microsomal triglyceride transfer protein inhibitor) and mipomersen (an antisense oligonucleotide inhibitor of ApoB-100 synthesis) can lower LDL-C by reducing ApoB synthesis,²¹ they are approved only in the small population of individuals with homozygous FH and liver toxicity and side effects are a concern.

Accordingly, current cholesterol management guidelines continue to offer LDL-C as the main target of lipid-modifying therapy, with statins as the primary treatment choice. The desire to build on statin therapy to prevent further progression of atherosclerosis and clinical CVD has encouraged continued focus on strategies to lower LDL-C to even greater extents.

Fortunately for practitioners, for the first time since lovastatin was approved in 1987, there is a new therapy approved by the FDA that significantly lowers LDL-C and, potentially, improves CV outcomes—the proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors. This review will focus on the PCSK9 inhibitors, a novel therapeutic class that reduces LDL-C through increased hepatic clearance. These drugs are rapidly emerging as an ideal adjunctive therapy to statins for patients at the highest risk and as a highly efficacious alternate therapy in patients intolerant of statins.

PCSK9 INHIBITORS: DISCOVERY, MECHANISM, AND THERAPEUTIC INTERVENTIONS

Two PCSK9 inhibitors have received FDA approval: alirocumab (Praluent) and evolocumab (Repatha). Among new molecular entities for clinical use, PCSK9 inhibitor therapies had one of the shortest durations from discovery to development and approval.

Mutations in the PCSK9 gene associated with autosomal dominant hypercholesterolemia were first identified in 2003 in a French family.²² The PCSK9 protein is now known to be a secreted enzymatic serine protease that is primarily synthesized in the liver and binds to the LDL receptor (LDL-R)/LDL-C complex on the surface of hepatocytes, marking the receptor for lysosomal degradation rather than recycling to the cell surface. Thus, it reduces the quantity of LDL-R that is available to remove LDL-C from circulation.²³ As a result, higher levels of PCSK9 are associated with higher levels of plasma LDL-C.

The clinical importance of PCSK9 in regulating LDL-C is supported by observed mutations and polymorphisms. Gain-of-function mutations that increase the activity of PCSK9 have been shown to be associated with elevated LDL-C, premature CVD, and myocardial infarction (MI).²⁴ Conversely, loss-of-function mutations (heterozygotes found in 1% to 3% of the population) result in decreased activity of PCSK9, lower LDL-C, and lower incidence of CVD (Table 1).^25–29 These observations, combined with data showing that homozygote loss-of-function individuals with very low LDL-C were generally very healthy, sparked interest in developing inhibition of PCSK9 activity as a therapeutic strategy for hyperlipidemia.

Multiple pharmacologic developments are aimed at inhibiting PCSK9, with many compounds in clinical trials. The approaches include gene silencing with loss-of-function mutations, synthetic peptides, oral small molecules, and monoclonal antibodies. Gene silencing was first observed in 2007 when administration of antisense oligonucleotides targeted to selectively inhibit PCSK9 mRNA was found to up-regulate LDL-R, thereby decreasing serum levels of LDL-C.³⁰

The first study to establish the role of synthetic peptides in PCSK9 inhibition was performed in 2008. In this study, the epidermal growth factor-like A synthetic peptide blocked the interaction between PCSK9 and LDL-R, thereby decreasing the degradation of LDL-R and preserving LDL uptake.³¹ Although studies are limited, synthetic peptides remain an area of great interest given their promising effects on lipid metabolism. Recently, a synthetic PCSK9-binding adnectin derived from the human fibronectin known as BMS-962476, had favorable results in a phase 1 clinical trial. An RNA interference molecule, subcutaneous ALN-PSC, inhibits PCSK9 gene expression by causing destruction of messenger RNA, thus inhibiting PCSK9 synthesis (Table 2).³²

PCSK9 INHIBITORS: CLINICAL TRIALS

Subcutaneously administered monoclonal antibodies targeting PCSK9 currently are the only PCSK9 inhibitors FDA-approved for clinical use. The first study to demonstrate efficacy in enhancing uptake of serum LDL-C was performed in 2009.³³ Multiple phase 1 and 2 studies soon followed, demonstrating acceptable safety and 50% to 70% reductions in LDL-C at upper-dose titrations.³⁴ Additionally, there were significant reductions in total cholesterol, ApoB, triglycerides, and lipoprotein(a).

These early developments paved the way for larger phase 3 trials (Table 3).^35–48 The PCSK9 inhibitors evolocumab and alirocumab have been shown in multiple phase 3 clinical trials to achieve a consistent dose-dependent 50% to 60% reduction in LDL-C across a broad range of CVD risk, pretreatment LDL-C levels, and background therapy: monotherapy (MENDEL-2, ODYSSEY COMBO I),^35,44 added to statin therapy (LAPLACE-2, ODYSSEY CHOICE I),^38,46 and in individuals with heterozygous FH (RUTHERFORD-2, ODYSSEY-FH).^37,42 Trials with bococizumab are under way.

The GAUSS-2 clinical trial (Goal Achievement after Utilizing an Anti-PCSK9 Antibody in Statin Intolerant Subjects-2) demonstrated similar efficacy in reducing LDL-C in patients with clinically assessed statin intolerance due to muscle-related adverse symptoms.³⁹ In GAUSS-3, patients were first identified as being statin intolerant secondary to muscle-associated symptoms based on a randomized, crossover trial of atorvastatin vs placebo.⁴⁰ The 43% of participants who experienced intolerable muscle-related symptoms on the statin but not on placebo were then randomized to evolocumab vs ezetimibe. Results showed significant reduction in LDL-C in the evolocumab group (52.8%) compared with the ezetimibe group (16.7%). Additionally, among patients with muscle symptoms on statin therapy, PCSK9 therapy was discontinued for muscle symptoms in only 0.7% of evolocumab recipients and 6.8% of ezetimibe recipients.

Overall, the PCSK9 inhibitors are generally well tolerated with injection site reactions being the most common side effect. A meta-analysis published in 2015 of 25 trials including more than 12,000 patients treated with evolocumab and alirocumab reported no significant difference in adverse events or safety outcomes vs placebo or ezetimibe.⁴⁹ Antidrug binding or neutralizing antibody production to these agents, thus far, has not been shown to be an issue. Additional analyses have not indicated an adverse effect on gonadal hormone levels or increased incidence of new-onset diabetes.

Two studies published in 2015 offer insight into longer term durability and safety as well as potential CVD outcome benefit (Table 4)^50,51:

• OSLER-1 and 2: Open-Label Study of Long-Term Evaluation against LDL-Cholesterol (OSLER) trials—evolocumab trial;⁵⁰
• ODYSSEY long term: Long-Term Safety and Tolerability of Alirocumab in High Cardiovascular Risk Patients with Hypercholesterolemia Not Adequately Controlled with Their Lipid Modifying Therapy—alirocumab trial.⁵¹

The OSLER trials reported durable LDL-C reductions of 61% and the ODYSSEY trial reported a LDL-C reduction of 62%.^50,51 In both studies, the overall occurrence of adverse events was similar to placebo, but both reported a higher rate of neurocognitive effects in the active treatment groups (evolocumab 0.9% vs 0.3% for standard therapy; alirocumab 1.2% vs 0.5% for placebo). It must be noted that although the absolute rate of neurocognitive adverse events is low, it is unclear if these events were related to the drugs themselves or to extreme lowering of LDL-C. Nevertheless, the FDA has raised concerns about neurocognitive events. A sub-study of the ongoing FOURIER trial with evolocumab—EBBINGHAUS—is expected to address this concern.

In addition, analyses of CV events showed that the PCSK9 inhibitors effectively cut the CV rate in half in both studies (Figure 1).^50,51 In the OSLER trials,⁵⁰ evolocumab recipients had 53% reduction in major CV events (0.95% vs 2.18% in the standard therapy group; P = .003). In ODYSSEY,⁵¹ alirocumab recipients had a 48% reduction in major CV events (1.7% vs 3.3% for placebo; P = .02). Furthermore, a 2015 meta-analysis of 24 phase 2 and 3 trials reported a statistically significant 55% reduction in all-cause mortality and 50% reduction in CV mortality with PCSK9 inhibitors.⁵²

For many reasons including short length of follow-up, study design, and small numbers of outcome events, the OSLER and ODYSSEY studies, although enticing, are exploratory and hypothesis-generating only and results need to be interpreted with caution. Nevertheless, they have set the stage for ongoing prospective randomized outcome trials that are studying the CV effects and tolerability of PCSK9 inhibitors over a longer time frame. These include the following trials.

• The Further Cardiovascular Outcomes Research with PCSK9 Inhibition in Subjects With Elevated Risk (FOURIER) is an ongoing trial with the primary end point of CV death, MI, hospitalization for unstable angina, stroke, or coronary revascularization in high-risk patients receiving evolocumab or placebo.⁵³
• The ODYSSEY trial is examining the effect of alirocumab vs placebo on the composite primary endpoint of coronary heart disease death, non-fatal MI, fatal and nonfatal ischemic stroke, and unstable angina requiring hospitalization in patients who have had an acute coronary syndrome event during the previous 4 to 52 weeks.⁵⁴
• The Evaluation of Bococizumab in Reducing the Occurrence of Major Cardiovascular Events in High Risk Subjects (SPIRE) trials are investigating the effect of bococizumab, a third PCSK9 “humanized” monoclonal antibody, vs placebo in reducing death, MI, stroke, or unstable angina in patients at high-risk of CVD who are receiving standard lipid-lowering therapy with LDL-C > 70 mg/dL (1.8 mmol/L) (SPIRE-1) or > 100 mg/dL (2.6 mmol/L) (SPIRE-2).^55,56  

Because these outcome trials are attempting to enroll more than 70,000 patients and are event driven, it is difficult to predict when they will be completed (Table 5).^53–56 However, recent estimates indicate completion of at least one trial by the end of 2016 or early 2017, with interim analyses of others expected at that time. It is hoped that they will answer the all-important question of whether PCSK9 inhibitors are associated with further CV event reduction benefit.

CURRENT FDA INDICATIONS AND GUIDELINES

The two PCSK9 inhibitors approved by the FDA—alirocumab (subcutaneous 75 mg every 2 weeks up titrated to 150 mg) and evolocumab (subcutaneous 140 mg every 2 weeks or 420 mg every 4 weeks)—are both indicated for use with statins in patients with heterozygous FH or known atherosclerotic CVD who require further reduction in LDL-C levels despite lifestyle interventions and use of maximally tolerated statins. Evolocumab has also been approved for use in patients with homozygous FH.

Although PCSK9 inhibitors are not specifically approved for patients unable to tolerate statins, the results of GAUSS-3, which documented that statin intolerance is a real, definable entity and very responsive to PCSK9 inhibition, makes these drugs promising agents for patients intolerant of statins and, thus, unable to benefit from high-intensity stain therapy.

In April 2016, the ACC released a clinical consensus update to their 2013 cholesterol guidelines, which is their first recommendation specifically addressing the use of non-statin therapies, including the newer PCSK9 inhibitors.⁵⁷ For high-risk patients with clinical atherosclerotic CVD or LDL-C > 190 and failure to achieve at least a 50% reduction in LDL-C on maximally tolerated statin, non-statins may be considered. Ezetimibe, given its safety and tolerability, should be the first additional medication added. Bile acid sequestrants may be used as a second-line therapy if ezetimibe is not tolerated and triglycerides are not elevated. If therapy goals are not met on maximally tolerated statin and ezetimibe, either approved PCSK9 inhibitor can be added or used to replace ezetimibe. The document also specifies that given the lack of long-term safety and efficacy data on the PCSK9 inhibitors, they are not recommended for use in primary prevention patients in the absence of FH.

CONCLUSION

Although statin therapy has been shown to substantially reduce LDL-C and CVD adverse events, there remains a high rate of inadequate goal achievement and residual CVD risk in patients receiving statins. Combination therapies with ezetimibe and resins to further lower LDL-C, fibrates and omega 3 fatty acids to lower triglycerides, and niacin to lower both and raise high-density-liproprotein cholesterol are available, even though additional CV risk reduction is minimal or elusive when these drugs are added to statin therapy.

The link between atherogenic lipoproteins and CVD is strong, and the need to develop therapies in addition to statins to substantially and safely reduce LDL-C remains a priority. The association of reduced PCSK9 activity with reduced LDL-C and CV events has led to rapid development and approval of monoclonal antibody therapies to inhibit PCSK9. In trials, these therapies have shown substantial and durable reductions in LDL-C of more than 50%, with acceptable tolerability. Now that PCSK9 inhibitors are approved by the FDA, extended data about long-term tolerability, safety, and efficacy and, most importantly, demonstration of additional reduction in CVD events are needed. It is hoped that the long-term ongoing trials will provide these data.

For the immediate future, statin therapy will continue to be the cornerstone of lipid and CVD risk management based on their low generic cost, proven CVD risk reduction, and clinicians’ comfort with their use. However, the reliable efficacy of PCSK9 inhibitors and the fact that statin therapy itself increases PCSK9 activity makes the addition of PCSK9 inhibitors to statins an attractive approach in high-risk patients failing to reach LDL-C treatment goals.

Although current indications are limited, there are patients at high CVD risk who would be appropriate candidates for these therapies. These include patients with the following:

• FH with lifetime burden of elevated LDL-C and associated low likelihood of achieving optimal LDL-C control on current available therapies
• Complete or partial statin intolerance with high-intensity statin dosing limited by side effects
• High CV risk who are not at LDL-C goal on current therapies.

Now that the first therapies are available, practitioners can expect newer approaches to tackle PCSK9-mediated LDL-C reduction. Bococizumab is lagging in phase 3 trials, but the SPIRE program is moving forward with special population studies expected to conclude in 2016 and simultaneous long-term outcomes trials. Other PCSK9 inhibitors being investigated include agents with more durable effect requiring less frequent injections, RNA-interference therapies, vaccinations, antisense therapies, and oral formulations.

The PCSK9 inhibitors hold promise as an adjunct to statin therapy. Their eventual clinical role will depend on a balance between substantial LDL-C reductions, long-term safety, tolerability, and reduction in CVD events vs the cost (estimated at $14,000 a year), access from payers, acceptance of injectable therapies, and magnitude of incremental benefit when added to current therapies. Nevertheless, initial clinical trial data are encouraging and these drugs may be an important addition to the therapeutic armamentarium against CVD.

Statin therapy has been shown to substantially reduce adverse events associated with low-density-lipoprotein cholesterol (LDL-C) and cardiovascular disease (CVD). Statins alone are often not adequate to achieve treatment goals, and residual CVD risk remains high. Combination therapies of statins with ezetimibe and resins to further lower LDL-C, fibrates and omega 3 fatty acids to lower triglycerides, and niacin to lower both and raise high-density-liproprotein cholesterol are available, but additional risk reduction has not been consistently demonstrated in clinical trials.

The link between atherogenic lipoproteins and CVD is strong, and the need to develop therapies in addition to statins to substantially and safely reduce LDL-C is a priority. The association of reduced proprotein convertase subtilisin/kexin type 9 (PCSK9) activity with reduced LDL-C and CVD events has led to the rapid development and approval of mono­clonal antibody therapies to inhibit PCSK9.

In this review, we discuss trials of these therapies that have shown durable reductions in LDL-C of more than 50%, with acceptable tolerability. Now that PCSK9 inhibitors are approved by the US Food and Drug Administration (FDA), extended data are needed as to long-term tolerability, safety, and efficacy of these agents and, most importantly, demonstration of additional reduction in CVD events.

A CASE FOR ADDITIONAL THERAPIES

CVD is the leading cause of morbidity and death in the United States, responsible for one in four deaths. Hyperlipidemia and, specifically, elevated LDL-C have been found to be important drivers of atherosclerosis and, in turn, adverse cardiovascular (CV) events. Likewise, numerous observational and clinical trials have shown that reducing LDL-C, particularly with statins, decreases CVD events.^1–4 More aggressive lowering with higher doses or more intensive statin therapy further reduces rates of adverse outcomes.^3,4 In addition, the pleiotropic effects of statins imply that not all of their benefits are derived from LDL-C lowering alone.⁵ Consequently, it is now standard practice to use statins at the highest tolerable dose to reach target LDL-C levels and prevent CV events in high-risk patients with CVD or multi­ple coronary artery disease risk factors, regardless of the LDL-C levels.^6,7

The American College of Cardiology (ACC) and the American Heart Association released cholesterol guidelines in 2013 that recommend a risk-based approach for statin therapy rather than targeting specific LDL-C levels.⁶ Although this evidence-based approach may better conform to clinical trials, the debate that lower LDL-C targets will further prevent CVD continues.

Indeed, it appears that lower is better, as demonstrated by the IMPROVE-IT trial.⁸ Although the control group receiving simvastatin monotherapy had low LDL-C levels (mean, 69.9 mg/dL; 1.8 mmol/L), the experimental group receiving simvastatin plus ezetimibe achieved even lower levels (mean, 53.2 mg/dL;1.4 mmol/L) after 1 year of therapy and had a significantly lower composite primary end point of CV death, major coronary event, or nonfatal stroke at 7 years (34.7% for simvastatin monotherapy vs 32.7% for combined therapy).⁹ Furthermore, the event-rate reduction with the addition of ezetimibe was the same as the average predicted by the Cholesterol Treatment Trialists’ meta-analysis: an LDL-C reduction of 1 mmol/L (38.6 mg/dL) yields a 23% risk reduction in major coronary events over 5 years.¹⁰ Although only a modest absolute reduction in outcomes, it supports the notion that further reduction of LDL-C levels by more potent therapies may offer greater benefit.

There is strong evidence that statin therapy reduces the risk of developing CVD in patients with or without a previous atherosclerotic event; however, residual CVD risk remains even for those on therapy. A contributing factor to this residual risk is that many statin-treated patients have insufficient response or intolerance and do not achieve adequate LDL-C reductions.

There are three clinically important patient populations who are inadequately managed with current therapies and remain at high risk of subsequent CV events; these are patients who would benefit from additional therapies.

1. Patients with familial hypercholesterolemia (FH). This is the most common genetic disorder in the world, yet it is frequently undiagnosed and untreated. Due to high baseline cholesterol levels, achieving LDL-C treatment goals is challenging.

• The prevalence may be closer to 1:200 to 1:250 rather than the often quoted 1:500.¹¹
• Fewer than 12% of patients with heterzygous FH achieve the LDL-C goal of < 100 mg/dL with maximal statin treatment alone or with a second agent.¹²

2. Patients with hyperlipidemia not due to FH who are at elevated CV risk and undertreated. In US and European surveys, between 50% and 60% of patients receiving statins with or without other therapies failed to reach LDL-C reduction goals.¹³

• Variation in response to statin treatment between individuals may be considerable.
• Poor adherence to statin therapy is common.

3. Patients with side effects to statins, particularly muscle symptoms that prevent statin use or substantially limit the dose.

• Although the incidence of myopathy is low (< 0.1%) and rhabdomyolysis is even less common, observational studies suggest that 10% to 20% of patients may limit statin use due to muscle-associated complaints including muscle aching, cramps, or weakness.¹⁴
• Side effects may be dose-dependent, limiting the use of the high-intensity statin doses that are frequently necessary to achieve LDL-C goals.

Consequently, there is great interest in developing therapies beyond statins that may further reduce CV events. However, treatments other than ezetimibe for further management of hyperlipidemia and risk reduction have failed to demonstrate consistent benefit when added to statin therapy.^15–19 The largest studies were with niacin and fibrates. Unfortunately, most trials demonstrated no overall outcomes benefit or only benefits in subgroup analyses, leaving the door open to other pharmacologic interventions.

Studies with the cholesterol ester transfer protein (CETP) inhibitor torcetrapib, in combination with statin therapy, actually demonstrated an overall increase in all-cause mortality in the treatment group.²⁰ Two large outcome trials of the CETP inhibitors dalcetrapib and evacetrapib were stopped after interim analysis predicted no benefit. Although drugs such as lomitapide (a microsomal triglyceride transfer protein inhibitor) and mipomersen (an antisense oligonucleotide inhibitor of ApoB-100 synthesis) can lower LDL-C by reducing ApoB synthesis,²¹ they are approved only in the small population of individuals with homozygous FH and liver toxicity and side effects are a concern.

Accordingly, current cholesterol management guidelines continue to offer LDL-C as the main target of lipid-modifying therapy, with statins as the primary treatment choice. The desire to build on statin therapy to prevent further progression of atherosclerosis and clinical CVD has encouraged continued focus on strategies to lower LDL-C to even greater extents.

Fortunately for practitioners, for the first time since lovastatin was approved in 1987, there is a new therapy approved by the FDA that significantly lowers LDL-C and, potentially, improves CV outcomes—the proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors. This review will focus on the PCSK9 inhibitors, a novel therapeutic class that reduces LDL-C through increased hepatic clearance. These drugs are rapidly emerging as an ideal adjunctive therapy to statins for patients at the highest risk and as a highly efficacious alternate therapy in patients intolerant of statins.

PCSK9 INHIBITORS: DISCOVERY, MECHANISM, AND THERAPEUTIC INTERVENTIONS

Two PCSK9 inhibitors have received FDA approval: alirocumab (Praluent) and evolocumab (Repatha). Among new molecular entities for clinical use, PCSK9 inhibitor therapies had one of the shortest durations from discovery to development and approval.

Mutations in the PCSK9 gene associated with autosomal dominant hypercholesterolemia were first identified in 2003 in a French family.²² The PCSK9 protein is now known to be a secreted enzymatic serine protease that is primarily synthesized in the liver and binds to the LDL receptor (LDL-R)/LDL-C complex on the surface of hepatocytes, marking the receptor for lysosomal degradation rather than recycling to the cell surface. Thus, it reduces the quantity of LDL-R that is available to remove LDL-C from circulation.²³ As a result, higher levels of PCSK9 are associated with higher levels of plasma LDL-C.

The clinical importance of PCSK9 in regulating LDL-C is supported by observed mutations and polymorphisms. Gain-of-function mutations that increase the activity of PCSK9 have been shown to be associated with elevated LDL-C, premature CVD, and myocardial infarction (MI).²⁴ Conversely, loss-of-function mutations (heterozygotes found in 1% to 3% of the population) result in decreased activity of PCSK9, lower LDL-C, and lower incidence of CVD (Table 1).^25–29 These observations, combined with data showing that homozygote loss-of-function individuals with very low LDL-C were generally very healthy, sparked interest in developing inhibition of PCSK9 activity as a therapeutic strategy for hyperlipidemia.

Multiple pharmacologic developments are aimed at inhibiting PCSK9, with many compounds in clinical trials. The approaches include gene silencing with loss-of-function mutations, synthetic peptides, oral small molecules, and monoclonal antibodies. Gene silencing was first observed in 2007 when administration of antisense oligonucleotides targeted to selectively inhibit PCSK9 mRNA was found to up-regulate LDL-R, thereby decreasing serum levels of LDL-C.³⁰

The first study to establish the role of synthetic peptides in PCSK9 inhibition was performed in 2008. In this study, the epidermal growth factor-like A synthetic peptide blocked the interaction between PCSK9 and LDL-R, thereby decreasing the degradation of LDL-R and preserving LDL uptake.³¹ Although studies are limited, synthetic peptides remain an area of great interest given their promising effects on lipid metabolism. Recently, a synthetic PCSK9-binding adnectin derived from the human fibronectin known as BMS-962476, had favorable results in a phase 1 clinical trial. An RNA interference molecule, subcutaneous ALN-PSC, inhibits PCSK9 gene expression by causing destruction of messenger RNA, thus inhibiting PCSK9 synthesis (Table 2).³²

PCSK9 INHIBITORS: CLINICAL TRIALS

Subcutaneously administered monoclonal antibodies targeting PCSK9 currently are the only PCSK9 inhibitors FDA-approved for clinical use. The first study to demonstrate efficacy in enhancing uptake of serum LDL-C was performed in 2009.³³ Multiple phase 1 and 2 studies soon followed, demonstrating acceptable safety and 50% to 70% reductions in LDL-C at upper-dose titrations.³⁴ Additionally, there were significant reductions in total cholesterol, ApoB, triglycerides, and lipoprotein(a).

These early developments paved the way for larger phase 3 trials (Table 3).^35–48 The PCSK9 inhibitors evolocumab and alirocumab have been shown in multiple phase 3 clinical trials to achieve a consistent dose-dependent 50% to 60% reduction in LDL-C across a broad range of CVD risk, pretreatment LDL-C levels, and background therapy: monotherapy (MENDEL-2, ODYSSEY COMBO I),^35,44 added to statin therapy (LAPLACE-2, ODYSSEY CHOICE I),^38,46 and in individuals with heterozygous FH (RUTHERFORD-2, ODYSSEY-FH).^37,42 Trials with bococizumab are under way.

The GAUSS-2 clinical trial (Goal Achievement after Utilizing an Anti-PCSK9 Antibody in Statin Intolerant Subjects-2) demonstrated similar efficacy in reducing LDL-C in patients with clinically assessed statin intolerance due to muscle-related adverse symptoms.³⁹ In GAUSS-3, patients were first identified as being statin intolerant secondary to muscle-associated symptoms based on a randomized, crossover trial of atorvastatin vs placebo.⁴⁰ The 43% of participants who experienced intolerable muscle-related symptoms on the statin but not on placebo were then randomized to evolocumab vs ezetimibe. Results showed significant reduction in LDL-C in the evolocumab group (52.8%) compared with the ezetimibe group (16.7%). Additionally, among patients with muscle symptoms on statin therapy, PCSK9 therapy was discontinued for muscle symptoms in only 0.7% of evolocumab recipients and 6.8% of ezetimibe recipients.

Overall, the PCSK9 inhibitors are generally well tolerated with injection site reactions being the most common side effect. A meta-analysis published in 2015 of 25 trials including more than 12,000 patients treated with evolocumab and alirocumab reported no significant difference in adverse events or safety outcomes vs placebo or ezetimibe.⁴⁹ Antidrug binding or neutralizing antibody production to these agents, thus far, has not been shown to be an issue. Additional analyses have not indicated an adverse effect on gonadal hormone levels or increased incidence of new-onset diabetes.

Two studies published in 2015 offer insight into longer term durability and safety as well as potential CVD outcome benefit (Table 4)^50,51:

• OSLER-1 and 2: Open-Label Study of Long-Term Evaluation against LDL-Cholesterol (OSLER) trials—evolocumab trial;⁵⁰
• ODYSSEY long term: Long-Term Safety and Tolerability of Alirocumab in High Cardiovascular Risk Patients with Hypercholesterolemia Not Adequately Controlled with Their Lipid Modifying Therapy—alirocumab trial.⁵¹

The OSLER trials reported durable LDL-C reductions of 61% and the ODYSSEY trial reported a LDL-C reduction of 62%.^50,51 In both studies, the overall occurrence of adverse events was similar to placebo, but both reported a higher rate of neurocognitive effects in the active treatment groups (evolocumab 0.9% vs 0.3% for standard therapy; alirocumab 1.2% vs 0.5% for placebo). It must be noted that although the absolute rate of neurocognitive adverse events is low, it is unclear if these events were related to the drugs themselves or to extreme lowering of LDL-C. Nevertheless, the FDA has raised concerns about neurocognitive events. A sub-study of the ongoing FOURIER trial with evolocumab—EBBINGHAUS—is expected to address this concern.

In addition, analyses of CV events showed that the PCSK9 inhibitors effectively cut the CV rate in half in both studies (Figure 1).^50,51 In the OSLER trials,⁵⁰ evolocumab recipients had 53% reduction in major CV events (0.95% vs 2.18% in the standard therapy group; P = .003). In ODYSSEY,⁵¹ alirocumab recipients had a 48% reduction in major CV events (1.7% vs 3.3% for placebo; P = .02). Furthermore, a 2015 meta-analysis of 24 phase 2 and 3 trials reported a statistically significant 55% reduction in all-cause mortality and 50% reduction in CV mortality with PCSK9 inhibitors.⁵²

For many reasons including short length of follow-up, study design, and small numbers of outcome events, the OSLER and ODYSSEY studies, although enticing, are exploratory and hypothesis-generating only and results need to be interpreted with caution. Nevertheless, they have set the stage for ongoing prospective randomized outcome trials that are studying the CV effects and tolerability of PCSK9 inhibitors over a longer time frame. These include the following trials.

• The Further Cardiovascular Outcomes Research with PCSK9 Inhibition in Subjects With Elevated Risk (FOURIER) is an ongoing trial with the primary end point of CV death, MI, hospitalization for unstable angina, stroke, or coronary revascularization in high-risk patients receiving evolocumab or placebo.⁵³
• The ODYSSEY trial is examining the effect of alirocumab vs placebo on the composite primary endpoint of coronary heart disease death, non-fatal MI, fatal and nonfatal ischemic stroke, and unstable angina requiring hospitalization in patients who have had an acute coronary syndrome event during the previous 4 to 52 weeks.⁵⁴
• The Evaluation of Bococizumab in Reducing the Occurrence of Major Cardiovascular Events in High Risk Subjects (SPIRE) trials are investigating the effect of bococizumab, a third PCSK9 “humanized” monoclonal antibody, vs placebo in reducing death, MI, stroke, or unstable angina in patients at high-risk of CVD who are receiving standard lipid-lowering therapy with LDL-C > 70 mg/dL (1.8 mmol/L) (SPIRE-1) or > 100 mg/dL (2.6 mmol/L) (SPIRE-2).^55,56  

Because these outcome trials are attempting to enroll more than 70,000 patients and are event driven, it is difficult to predict when they will be completed (Table 5).^53–56 However, recent estimates indicate completion of at least one trial by the end of 2016 or early 2017, with interim analyses of others expected at that time. It is hoped that they will answer the all-important question of whether PCSK9 inhibitors are associated with further CV event reduction benefit.

CURRENT FDA INDICATIONS AND GUIDELINES

The two PCSK9 inhibitors approved by the FDA—alirocumab (subcutaneous 75 mg every 2 weeks up titrated to 150 mg) and evolocumab (subcutaneous 140 mg every 2 weeks or 420 mg every 4 weeks)—are both indicated for use with statins in patients with heterozygous FH or known atherosclerotic CVD who require further reduction in LDL-C levels despite lifestyle interventions and use of maximally tolerated statins. Evolocumab has also been approved for use in patients with homozygous FH.

Although PCSK9 inhibitors are not specifically approved for patients unable to tolerate statins, the results of GAUSS-3, which documented that statin intolerance is a real, definable entity and very responsive to PCSK9 inhibition, makes these drugs promising agents for patients intolerant of statins and, thus, unable to benefit from high-intensity stain therapy.

In April 2016, the ACC released a clinical consensus update to their 2013 cholesterol guidelines, which is their first recommendation specifically addressing the use of non-statin therapies, including the newer PCSK9 inhibitors.⁵⁷ For high-risk patients with clinical atherosclerotic CVD or LDL-C > 190 and failure to achieve at least a 50% reduction in LDL-C on maximally tolerated statin, non-statins may be considered. Ezetimibe, given its safety and tolerability, should be the first additional medication added. Bile acid sequestrants may be used as a second-line therapy if ezetimibe is not tolerated and triglycerides are not elevated. If therapy goals are not met on maximally tolerated statin and ezetimibe, either approved PCSK9 inhibitor can be added or used to replace ezetimibe. The document also specifies that given the lack of long-term safety and efficacy data on the PCSK9 inhibitors, they are not recommended for use in primary prevention patients in the absence of FH.

CONCLUSION

Although statin therapy has been shown to substantially reduce LDL-C and CVD adverse events, there remains a high rate of inadequate goal achievement and residual CVD risk in patients receiving statins. Combination therapies with ezetimibe and resins to further lower LDL-C, fibrates and omega 3 fatty acids to lower triglycerides, and niacin to lower both and raise high-density-liproprotein cholesterol are available, even though additional CV risk reduction is minimal or elusive when these drugs are added to statin therapy.

The link between atherogenic lipoproteins and CVD is strong, and the need to develop therapies in addition to statins to substantially and safely reduce LDL-C remains a priority. The association of reduced PCSK9 activity with reduced LDL-C and CV events has led to rapid development and approval of monoclonal antibody therapies to inhibit PCSK9. In trials, these therapies have shown substantial and durable reductions in LDL-C of more than 50%, with acceptable tolerability. Now that PCSK9 inhibitors are approved by the FDA, extended data about long-term tolerability, safety, and efficacy and, most importantly, demonstration of additional reduction in CVD events are needed. It is hoped that the long-term ongoing trials will provide these data.

For the immediate future, statin therapy will continue to be the cornerstone of lipid and CVD risk management based on their low generic cost, proven CVD risk reduction, and clinicians’ comfort with their use. However, the reliable efficacy of PCSK9 inhibitors and the fact that statin therapy itself increases PCSK9 activity makes the addition of PCSK9 inhibitors to statins an attractive approach in high-risk patients failing to reach LDL-C treatment goals.

Although current indications are limited, there are patients at high CVD risk who would be appropriate candidates for these therapies. These include patients with the following:

• FH with lifetime burden of elevated LDL-C and associated low likelihood of achieving optimal LDL-C control on current available therapies
• Complete or partial statin intolerance with high-intensity statin dosing limited by side effects
• High CV risk who are not at LDL-C goal on current therapies.

Now that the first therapies are available, practitioners can expect newer approaches to tackle PCSK9-mediated LDL-C reduction. Bococizumab is lagging in phase 3 trials, but the SPIRE program is moving forward with special population studies expected to conclude in 2016 and simultaneous long-term outcomes trials. Other PCSK9 inhibitors being investigated include agents with more durable effect requiring less frequent injections, RNA-interference therapies, vaccinations, antisense therapies, and oral formulations.

The PCSK9 inhibitors hold promise as an adjunct to statin therapy. Their eventual clinical role will depend on a balance between substantial LDL-C reductions, long-term safety, tolerability, and reduction in CVD events vs the cost (estimated at $14,000 a year), access from payers, acceptance of injectable therapies, and magnitude of incremental benefit when added to current therapies. Nevertheless, initial clinical trial data are encouraging and these drugs may be an important addition to the therapeutic armamentarium against CVD.

On November 12, 2013, a joint task force for the American College of Cardiology and American Heart Association released new guidelines for treating high blood cholesterol to reduce the risk of atherosclerotic cardiovascular disease (ASCVD) in adults.¹

This document arrives after several years of intense deliberation, 12 years after the third Adult Treatment Panel (ATP III) guidelines,² and 8 years after an ATP III update recommending that low-density lipoprotein cholesterol (LDL-C) levels be lowered aggressively (to less than 70 mg/dL) as an option in patients at high risk.³ It represents a major shift in the approach to and management of high blood cholesterol and has sparked considerable controversy.

In the following commentary, we summarize the new guidelines and the philosophy employed by the task force in generating them. We will also examine some advantages and what we believe to be several shortcomings of the new guidelines. These latter points are illustrated through case examples.

IN RANDOMIZED CONTROLLED TRIALS WE TRUST

In collaboration with the National Heart, Lung, and Blood Institute of the National Institutes of Health, the American College of Cardiology and American Heart Association formed an expert panel task force in 2008.

The task force elected to use only evidence from randomized controlled trials, systematic reviews, and meta-analyses of randomized controlled trials (and only predefined outcomes of the trials, not post hoc analyses) in formulating its recommendations, with the goal of providing the strongest possible evidence.

The authors state that “By using [randomized controlled trial] data to identify those most likely to benefit [emphasis in original] from cholesterol-lowering statin therapy, the recommendations will be of value to primary care clinicians as well as specialists concerned with ASCVD prevention. Importantly, the recommendations were designed to be easy to use in the clinical setting, facilitating the implementation of a strategy of risk assessment and treatment focused on the prevention of ASCVD.”³ They also state the guidelines are meant to “inform clinical judgment, not replace it” and that clinician judgment in addition to discussion with patients remains vital.

During the deliberations, the National Heart, Lung, and Blood Institute removed itself from participating, stating its mission no longer included drafting new guidelines. Additionally, other initial members of the task force removed themselves because of disagreement and concerns about the direction of the new guidelines.

These guidelines, and their accompanying new cardiovascular risk calculator,⁴ were released without a preliminary period to allow for open discussion, comment, and critique by physicians outside the panel. No attempt was made to harmonize the guidelines with previous versions (eg, ATP III) or with current international guidelines.

WHAT’S NEW IN THE GUIDELINES?

The following are the major changes in the new guidelines for treating high blood cholesterol:

• Treatment goals for LDL-C and non-high-density lipoprotein cholesterol (non-HDL-C) are no longer recommended.
• High-intensity and moderate-intensity statin treatment is emphasized, and low-intensity statin therapy is nearly eliminated.
• “ASCVD” now includes stroke in addition to coronary heart disease and peripheral arterial disease.
• Four groups are targeted for treatment (see below).
• Nonstatin therapies have been markedly de-emphasized.
• No guidelines are provided for treating high triglyceride levels.

The new guidelines emphasize lifestyle modification as the foundation for reducing risk, regardless of cholesterol therapy. No recommendations are given for patients with New York Heart Association class II, III, or IV heart failure or for hemodialysis patients, because there were insufficient data from randomized controlled trials to support recommendations. Similarly, the guidelines apply only to people between the ages of 40 and 75 (risk calculator ages 40–79), because the authors believed there was not enough evidence from randomized controlled trials to allow development of guidelines outside of this age range.

FOUR MAJOR STATIN TREATMENT GROUPS

The new guidelines specify four groups that merit intensive or moderately intensive statin therapy (Table 1)¹:

• People with clinical ASCVD
• People with LDL-C levels of 190 mg/dL or higher
• People with diabetes, age 40 to 75
• People without diabetes, age 40 to 75, with LDL-C levels 70–189 mg/dL, and a 10-year ASCVD risk of 7.5% or higher as determined by the new risk calculator⁴ (which also calculates the lifetime risk of ASCVD).

Below, we will address each of these four groups and provide case scenarios to consider. In general, our major disagreements with the new recommendations pertain to the first and fourth categories.

GROUP 1: PEOPLE WITH CLINICAL ASCVD

Advantages of the new guidelines

• They appropriately recommend statins in the highest tolerated doses as first-line treatment for this group at high risk.
• They designate all patients with ASCVD, including those with coronary, peripheral, and cerebrovascular disease, as a high-risk group.
• Without target LDL-C levels, treatment is simpler than before, requiring less monitoring of lipid levels. (This can also be seen as a limitation, as we discuss below.)

Limitations of the new guidelines

• They make follow-up LDL-C levels irrelevant, seeming to assume that there is no gradation in residual risk and, thus, no need to tailor therapy to the individual.
• Patients no longer have a goal to strive for or a way to monitor their progress.
• The guidelines ignore the pathophysiology of coronary artery disease and evidence of residual risk in patients on both moderate-intensity and high-intensity statin therapy.
• They also ignore the potential benefits of treating to lower LDL-C or non-HDL-C goals, thus eliminating consideration of multidrug therapy. They do not address patients with recurrent cardiovascular events already on maximal tolerated statin doses.
• They undermine the potential development and use of new therapies for dysplipidemia in patients with ASCVD.

Case 1: Is LDL-C 110 mg/dL low enough?

A 52-year-old African American man presents with newly discovered moderate coronary artery disease that is not severe enough to warrant stenting. He has no history of hypertension, diabetes mellitus, or smoking. His systolic blood pressure is 130 mm Hg, and his body mass index is 26 kg/m². He exercises regularly and follows a low-cholesterol diet. He has the following fasting lipid values:

• Total cholesterol 290 mg/dL
• HDL-C 50 mg/dL
• Triglycerides 250 mg/dL
• Calculated LDL-C 190 mg/dL.

Two months later, after beginning atorvastatin 80 mg daily, meeting with a nutritionist, and redoubling his dietary efforts, his fasting lipid concentrations are:

• Total cholesterol 180 mg/dL
• HDL-C 55 mg/dL
• Triglycerides 75 mg/dL
• Calculated LDL-C 110 mg/dL.

Comment: Lack of LDL-C goals is a flaw

The new guidelines call for patients with known ASCVD, such as this patient, to receive intensive statin therapy—which he got.

However, once a patient is on therapy, the new guidelines do not encourage repeating the lipid panel other than to assess compliance. With intensive therapy, we expect a reduction in LDL-C of at least 50% (Table 1), but patient-to-patient differences in response to medications are common, and without repeat testing we would have no way of gauging this patient’s residual risk.

Further, the new guidelines emphasize the lack of hard outcome data supporting the addition of another lipid-lowering drug to a statin, although they do indicate that one can consider it. In a patient at high risk, such as this one, would you be comfortable with an LDL-C value of 110 mg/dL on maximum statin therapy? Would you consider adding a nonstatin drug?

A preponderance of data shows that LDL plays a causal role in ASCVD development and adverse events. Genetic data show that the LDL particle and the LDL receptor pathway are mechanistically linked to ASCVD pathogenesis, with lifetime exposure as a critical determinant of risk.^5,6 Moreover, randomized controlled trials of statins and other studies of cholesterol-lowering show a reproducible relationship between the LDL-C level achieved and absolute risk (Figure 1).^7–24 We believe the totality of data constitutes a strong rationale for targeting LDL-C and establishing goals for lowering its levels. For these reasons, we believe that removing LDL-C goals is a fundamental flaw of the new guidelines.

The reason for the lack of data from randomized controlled trials demonstrating benefits of adding therapies to statins (when LDL-C is still high) or benefits of treating to specific goals is that no such trials have been performed. Even trials of nonpharmacologic means of lowering LDL-C, such as ileal bypass, which was used in the Program on the Surgical Control of the Hyperlipidemias trial,²⁰ provide independent evidence that lowering LDL-C reduces the risk of ASCVD (Figure 1).

In addition, trials of nonstatin drugs, such as the Coronary Drug Project,25 which tested niacin, also showed outcome benefits. On the other hand, studies such as the Atherothrombosis Intervention in Metabolic Syndrome With Low HDL/High Triglycerides: Impact on Global Health26 and Treatment of HDL to Reduce the Incidence of Vascular Events27 trials did not show additional risk reduction when niacin was added to statin therapy. However, the study designs arguably had flaws, including requirement of aggressive LDL-lowering with statins, with LDL-C levels below 70 to 80 mg/dL before randomization.

Therefore, these trials do not tell us what to do for a patient on maximal intensive therapy who has recurrent ASCVD events or who, like our patient, has an LDL-C level higher than previous targets.

For this patient, we would recommend adding a second medication to further lower his LDL-C, but discussing with him the absence of proven benefit in clinical trials and the risks of side effects. At present, lacking LDL-C goals in the new guidelines, we are keeping with the ATP III goals to help guide therapeutic choices and individualize patient management.

GROUP 2: PEOPLE WITH LDL-C ≥ 190

Advantages of the new guidelines

• They state that these patients should receive statins in the highest tolerated doses, which is universally accepted.

Limitations of the new guidelines

• The new guidelines mention only that one “may consider” adding a second agent if LDL-C remains above 190 mg/dL after maximum-dose therapy. Patients with familial hypercholesterolemia or other severe forms of hypercholesterolemia typically end up on multidrug therapy to further reduce LDL-C. The absence of randomized controlled trial data in this setting to show an additive value of second and third lipid-lowering agents does not mean these agents do not provide benefit.

GROUP 3: DIABETES, AGE 40–75, LDL-C 70–189, NO CLINICAL ASCVD

Advantages of the new guidelines

• They call for aggressive treatment of people with diabetes, a group at high risk that derives significant benefit from statin therapy, as shown in randomized controlled trials.

Limitations of the new guidelines

• Although high-intensity statin therapy is indicated for this group, we believe that, using the new risk calculator, some patients may receive overly aggressive treatment, thus increasing the possibility of statin side effects.
• The guidelines do not address patients younger than 40 or older than 75.
• Diabetic patients have a high residual risk of ASCVD events, even on statin therapy. Yet the guidelines ignore the potential benefits of more aggressive LDL-lowering or non-LDL secondary targets for therapy.

Case 2: How low is too low?

A 63-year-old white woman, a nonsmoker with recently diagnosed diabetes, is seen by her primary care physician. She has hypertension, for which she takes lisinopril 5 mg daily. Her fasting lipid values are:

• Total cholesterol 160 mg/dL
• HDL-C 64 mg/dL
• Triglycerides 100 mg/dL
• Calculated LDL-C 76 mg/dL.

Her systolic blood pressure is 129 mm Hg, and based on the new risk calculator, her 10-year risk of cardiovascular disease is 10.2%. According to the new guidelines, she should be started on high-intensity statin treatment (Table 1).

Although this is an acceptable initial course of action, it necessitates close vigilance, since it may actually drive her LDL-C level too low. Randomized controlled trials have typically used an LDL-C concentration of less than or equal to 25 mg/dL as the safety cutoff. With a typical LDL-C reduction of at least 50% on high-intensity statins, our patient’s expected LDL-C level will likely be in the low 30s. We believe this would be a good outcome, provided that she tolerates the medication without adverse effects. However, responses to statins vary from patient to patient.

High-intensity statin therapy may not be necessary to reduce risk adequately in all patients who have diabetes without preexisting vascular disease. The Collaborative Atorvastatin Diabetes Study¹² compared atorvastatin 10 mg vs placebo in people with type 2 diabetes, age 40 to 75, who had one or more cardiovascular risk factors but no signs or symptoms of preexisting ASCVD and who had only average or below-average cholesterol levels—precisely like this patient. The trial was terminated early because of a clear benefit (a 37% reduction in the composite end point of major adverse cardiovascular events) in the intervention group. For our patient, we believe an alternative and acceptable approach would be to begin moderate-intensity statin therapy (eg, with atorvastatin 10 mg) (Table 1).

Alternatively, in a patient with diabetes and previous atherosclerotic vascular disease or with a high 10-year risk and high LDL-C, limiting treatment to high-intensity statin therapy by itself may deny them the potential benefits of combination therapies and targeting to lower LDL-C levels or non-HDL-C secondary targets. Guidelines from the American Diabetes Association²⁸ and the American Association of Clinical Endocrinologists²⁹ continue to recommend an LDL-C goal of less than 70 mg/dL in patients at high risk, a non-HDL-C less than 100 mg/dL, an apolipoprotein B less than 80 mg/dL, and an LDL particle number less than 1,000 nmol/L.

GROUP 4: AGE 40–75, LDL-C 70–189, NO ASCVD, BUT 10-YEAR RISK ≥ 7.5%

Advantages of the new guidelines

• They may reduce ASCVD events for patients at higher risk.
• The risk calculator is easy to use and focuses on global risk, ie, all forms of ASCVD.
• The guidelines promote discussion of risks and benefits between patients and providers.

Limitations of the new guidelines

• The new risk calculator is controversial (see below).
• There is potential for overtreatment, particularly in older patients.
• There is potential for undertreatment, particularly in patients with an elevated LDL-C but whose 10-year risk is less than 7.5% because they are young.
• The guidelines do not address patients younger than 40 or older than 75.
• They do not take into account some traditional risk factors, such as family history, and nontraditional risk factors such as C-reactive protein as measured by ultrasensitive assays, lipoprotein(a), and apolipoprotein B.

Risk calculator controversy

The new risk calculator has aroused strong opinions on both sides of the aisle.

Shortly after the new guidelines were released, cardiologists Dr. Paul Ridker and Dr. Nancy Cook from Brigham and Women’s Hospital in Boston published analyses³⁰ showing that the new risk calculator, which was based on older data from several large cohorts such as the Atherosclerosis Risk in Communities study,³¹ the Cardiovascular Health Study,³² the Coronary Artery Risk Development in Young Adults study,³³ and the Framingham Heart Study,^34,35 was inaccurate in other cohorts. Specifically, in more-recent cohorts (the Women’s Health Study,³⁶ Physicians’ Health Study,³⁷ and Women’s Health Initiative³⁸), the new calculator overestimates the 10-year risk of ASCVD by 75% to 150%.³⁰ Using the new calculator would make approximately 30 million more Americans eligible for statin treatment. The concern is that patients at lower risk would be treated and exposed to potential side effects of statin therapy.

In addition, the risk calculator relies heavily on age and sex and does not include other factors such as triglyceride level, family history, C-reactive protein, or lipoprotein(a). Importantly, and somewhat ironically given the otherwise absolute adherence to randomized controlled trial data for guideline development, the risk calculator has never been verified in prospective studies to adequately show that using it reduces ASCVD events.

Based on the risk calculator, essentially any African American man in his early 60s with no other risk factors has a 10-year risk of ASCVD of 7.5% or higher and, according to the new guidelines, should receive at least moderate-intensity statin therapy.

For example, consider a 64-year-old African American man whose systolic blood pressure is 129 mm Hg, who does not smoke, does not have diabetes, and does not have hypertension, and whose total cholesterol level is 180 mg/dL, HDL-C 70 mg/dL, triglycerides 130 mg/dL, and calculated LDL-C 84 mg/dL. His calculated 10-year risk is, surprisingly, 7.5%.

Alternatively, his twin brother is a two-pack-per-day smoker with untreated hypertension and systolic blood pressure 150 mm Hg, with fasting total cholesterol 153 mg/dL, HDL-C 70 mg/dL, triglycerides 60 mg/dL, and LDL-C 71 mg/dL. His calculated 10-year risk is 10.5%, so according to the new guidelines, he too should receive high-intensity statin therapy. Yet this patient clearly needs better blood pressure control and smoking cessation as his primary risk-reduction efforts, not a statin. While assessing global risk is important, a shortcoming of the new guidelines is that they can inappropriately lead to treating the risk score, not individualizing the treatment to the patient. Because of the errors inherent in the risk calculator, some experts have called for a temporary halt on implementing the new guidelines until the risk calculator can be further validated. In November 2013, the American Heart Association and the American College of Cardiology reaffirmed their support of the new guidelines and recommended that they be implemented as planned. As of the time this manuscript goes to print, there are no plans to halt implementation of the new guidelines.

Case 4: Undertreating a primary prevention patient

A 25-year-old white man with no medical history has a total cholesterol level of 310 mg/dL, HDL-C 50 mg/dL, triglycerides 400 mg/dL, and calculated LDL-C 180 mg/dL. He does not smoke or have hypertension or diabetes but has a strong family history of premature coronary disease (his father died of myocardial infarction at age 42). His body mass index is 25 kg/m². Because he is less than 40 years old, the risk calculator does not apply to him.

If we assume he remains untreated and returns at age 40 with the same clinical factors and laboratory values, his calculated 10-year risk of an ASCVD event according to the new risk calculator will still be only 3.1%. Assuming his medical history remains unchanged as he continues to age, his 10-year risk would not reach 7.5% until he is 58. Would you feel comfortable waiting 33 years before starting statin therapy in this patient?

Waiting for dyslipidemic patients to reach middle age before starting LDL-C-lowering therapy is a failure of prevention. For practical reasons, there are no data from randomized controlled trials with hard outcomes in younger people. Nevertheless, a tenet of preventive cardiology is that cumulative exposure accelerates the “vascular age” ahead of the chronological age. This case illustrates why individualized recommendations guided by LDL-C goals as a target for therapy are needed. For this 25-year-old patient, we would recommend starting an intermediate- or high-potency statin.

Case 5: Rheumatoid arthritis

A 60-year-old postmenopausal white woman with severe rheumatoid arthritis presents for cholesterol evaluation. Her total cholesterol level is 235 mg/dL, HDL-C 50 mg/dL, and LDL-C 165 mg/dL. She does not smoke or have hypertension or diabetes. Her systolic blood pressure is 110 mm Hg. She has elevated C-reactive protein on an ultrasensitive assay and elevated lipoprotein(a).

Her calculated 10-year risk of ASCVD is 3.0%. Assuming her medical history remains the same, she would not reach a calculated 10-year risk of at least 7.5% until age 70. We suggest starting moderate- or high-dose statin therapy in this case, based on data (not from randomized controlled trials) showing an increased risk of ASCVD events in patients with rheumatologic disease, increased lipoprotein(a), and inflammatory markers like C-reactive protein. However, the current guidelines do not address this scenario, other than to suggest that clinician consideration can be given to other risk markers such as these, and that these findings should be discussed in detail with the patient. The Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin trial¹⁶ showed a dramatic ASCVD risk reduction in just such patients (Figure 1).

APPLAUSE—AND RESERVATIONS

The newest guidelines for treating high blood cholesterol represent a monumental shift away from using target levels of LDL-C and non-HDL-C and toward a focus on statin intensity for patients in the four highest-risk groups.

We applaud the expert panel for its idealistic approach of using only data from randomized controlled trials, for placing more emphasis on higher-intensity statin treatment, for including stroke in the new definition of ASCVD, and for focusing more attention on treating diabetic patients more aggressively. Simplifying the guidelines is a noble goal. Emphasizing moderate-to-high-intensity statin therapy in patients at moderate-to-high risk should have substantial long-term public health benefits.

However, as we have shown in the case examples, there are significant limitations, and some patients can end up being overtreated, while others may be undertreated.

Guidelines need to be crafted by looking at all the evidence, including the pathophysiology of the disease process, not just data from randomized controlled trials. It is difficult to implement a guideline that on one hand used randomized controlled trials exclusively for recommendations, but on the other hand used an untested risk calculator to guide therapy. Randomized controlled trials are not available for every scenario.

Further, absence of randomized controlled trial data in a given scenario should not be interpreted as evidence of lack of benefit. An example of this is a primary-prevention patient under age 40 with elevated LDL-C below the 190 mg/dL cutoff who otherwise is healthy and without risk factors (eg, Case 4). By disregarding all evidence that is not from randomized controlled trials, the expert panel fails to account for the extensive pathophysiology of ASCVD, which often begins at a young age and takes decades to develop.^5,6,39 An entire generation of patients who have not reached the age of inclusion in most randomized controlled trials with hard outcomes is excluded (unless the LDL-C level is very high), potentially setting back decades of progress in the field of prevention. Prevention only works if started. With childhood and young adult obesity sharply rising, we should not fail to address the under-40-year-old patient population in our guidelines.

Guidelines are designed to be expert opinion, not to dictate practice. Focusing on the individual patient instead of the general population at risk, the expert panel appropriately emphasizes the “importance of clinician judgment, weighing potential benefits, adverse effects, drug-drug interactions and patient preferences.” However, by excluding all data that do not come from randomized controlled trials, the panel neglects a very large base of knowledge and leaves many clinicians without as much expert opinion as we had hoped for.

LDL-C goals are important: they provide a scorecard to help the patient with lifestyle and dietary changes. They provide the health care provider guidance in making treatment decisions and focusing on treatment of a single patient, not a population. Moreover, if a patient has difficulty taking standard doses of statins because of side effects, the absence of LDL-C goals makes decision-making nearly impossible. We hope physicians will rely on LDL-C goals in such situations, falling back on the ATP III recommendations, although many patients may simply go untreated until they present with ASCVD or until they “age in” to a higher risk category.

We suggest caution in strict adherence to the new guidelines and instead urge physicians to consider a hybrid of the old guidelines (using the ATP III LDL-C goals) and the new ones (emphasizing global risk assessment and high-intensity statin treatment).

On November 12, 2013, a joint task force for the American College of Cardiology and American Heart Association released new guidelines for treating high blood cholesterol to reduce the risk of atherosclerotic cardiovascular disease (ASCVD) in adults.¹

This document arrives after several years of intense deliberation, 12 years after the third Adult Treatment Panel (ATP III) guidelines,² and 8 years after an ATP III update recommending that low-density lipoprotein cholesterol (LDL-C) levels be lowered aggressively (to less than 70 mg/dL) as an option in patients at high risk.³ It represents a major shift in the approach to and management of high blood cholesterol and has sparked considerable controversy.

In the following commentary, we summarize the new guidelines and the philosophy employed by the task force in generating them. We will also examine some advantages and what we believe to be several shortcomings of the new guidelines. These latter points are illustrated through case examples.

IN RANDOMIZED CONTROLLED TRIALS WE TRUST

In collaboration with the National Heart, Lung, and Blood Institute of the National Institutes of Health, the American College of Cardiology and American Heart Association formed an expert panel task force in 2008.

The task force elected to use only evidence from randomized controlled trials, systematic reviews, and meta-analyses of randomized controlled trials (and only predefined outcomes of the trials, not post hoc analyses) in formulating its recommendations, with the goal of providing the strongest possible evidence.

The authors state that “By using [randomized controlled trial] data to identify those most likely to benefit [emphasis in original] from cholesterol-lowering statin therapy, the recommendations will be of value to primary care clinicians as well as specialists concerned with ASCVD prevention. Importantly, the recommendations were designed to be easy to use in the clinical setting, facilitating the implementation of a strategy of risk assessment and treatment focused on the prevention of ASCVD.”³ They also state the guidelines are meant to “inform clinical judgment, not replace it” and that clinician judgment in addition to discussion with patients remains vital.

During the deliberations, the National Heart, Lung, and Blood Institute removed itself from participating, stating its mission no longer included drafting new guidelines. Additionally, other initial members of the task force removed themselves because of disagreement and concerns about the direction of the new guidelines.

These guidelines, and their accompanying new cardiovascular risk calculator,⁴ were released without a preliminary period to allow for open discussion, comment, and critique by physicians outside the panel. No attempt was made to harmonize the guidelines with previous versions (eg, ATP III) or with current international guidelines.

WHAT’S NEW IN THE GUIDELINES?

The following are the major changes in the new guidelines for treating high blood cholesterol:

• Treatment goals for LDL-C and non-high-density lipoprotein cholesterol (non-HDL-C) are no longer recommended.
• High-intensity and moderate-intensity statin treatment is emphasized, and low-intensity statin therapy is nearly eliminated.
• “ASCVD” now includes stroke in addition to coronary heart disease and peripheral arterial disease.
• Four groups are targeted for treatment (see below).
• Nonstatin therapies have been markedly de-emphasized.
• No guidelines are provided for treating high triglyceride levels.

The new guidelines emphasize lifestyle modification as the foundation for reducing risk, regardless of cholesterol therapy. No recommendations are given for patients with New York Heart Association class II, III, or IV heart failure or for hemodialysis patients, because there were insufficient data from randomized controlled trials to support recommendations. Similarly, the guidelines apply only to people between the ages of 40 and 75 (risk calculator ages 40–79), because the authors believed there was not enough evidence from randomized controlled trials to allow development of guidelines outside of this age range.

FOUR MAJOR STATIN TREATMENT GROUPS

The new guidelines specify four groups that merit intensive or moderately intensive statin therapy (Table 1)¹:

• People with clinical ASCVD
• People with LDL-C levels of 190 mg/dL or higher
• People with diabetes, age 40 to 75
• People without diabetes, age 40 to 75, with LDL-C levels 70–189 mg/dL, and a 10-year ASCVD risk of 7.5% or higher as determined by the new risk calculator⁴ (which also calculates the lifetime risk of ASCVD).

Below, we will address each of these four groups and provide case scenarios to consider. In general, our major disagreements with the new recommendations pertain to the first and fourth categories.

GROUP 1: PEOPLE WITH CLINICAL ASCVD

Advantages of the new guidelines

• They appropriately recommend statins in the highest tolerated doses as first-line treatment for this group at high risk.
• They designate all patients with ASCVD, including those with coronary, peripheral, and cerebrovascular disease, as a high-risk group.
• Without target LDL-C levels, treatment is simpler than before, requiring less monitoring of lipid levels. (This can also be seen as a limitation, as we discuss below.)

Limitations of the new guidelines

• They make follow-up LDL-C levels irrelevant, seeming to assume that there is no gradation in residual risk and, thus, no need to tailor therapy to the individual.
• Patients no longer have a goal to strive for or a way to monitor their progress.
• The guidelines ignore the pathophysiology of coronary artery disease and evidence of residual risk in patients on both moderate-intensity and high-intensity statin therapy.
• They also ignore the potential benefits of treating to lower LDL-C or non-HDL-C goals, thus eliminating consideration of multidrug therapy. They do not address patients with recurrent cardiovascular events already on maximal tolerated statin doses.
• They undermine the potential development and use of new therapies for dysplipidemia in patients with ASCVD.

Case 1: Is LDL-C 110 mg/dL low enough?

A 52-year-old African American man presents with newly discovered moderate coronary artery disease that is not severe enough to warrant stenting. He has no history of hypertension, diabetes mellitus, or smoking. His systolic blood pressure is 130 mm Hg, and his body mass index is 26 kg/m². He exercises regularly and follows a low-cholesterol diet. He has the following fasting lipid values:

• Total cholesterol 290 mg/dL
• HDL-C 50 mg/dL
• Triglycerides 250 mg/dL
• Calculated LDL-C 190 mg/dL.

Two months later, after beginning atorvastatin 80 mg daily, meeting with a nutritionist, and redoubling his dietary efforts, his fasting lipid concentrations are:

• Total cholesterol 180 mg/dL
• HDL-C 55 mg/dL
• Triglycerides 75 mg/dL
• Calculated LDL-C 110 mg/dL.

Comment: Lack of LDL-C goals is a flaw

The new guidelines call for patients with known ASCVD, such as this patient, to receive intensive statin therapy—which he got.

However, once a patient is on therapy, the new guidelines do not encourage repeating the lipid panel other than to assess compliance. With intensive therapy, we expect a reduction in LDL-C of at least 50% (Table 1), but patient-to-patient differences in response to medications are common, and without repeat testing we would have no way of gauging this patient’s residual risk.

Further, the new guidelines emphasize the lack of hard outcome data supporting the addition of another lipid-lowering drug to a statin, although they do indicate that one can consider it. In a patient at high risk, such as this one, would you be comfortable with an LDL-C value of 110 mg/dL on maximum statin therapy? Would you consider adding a nonstatin drug?

A preponderance of data shows that LDL plays a causal role in ASCVD development and adverse events. Genetic data show that the LDL particle and the LDL receptor pathway are mechanistically linked to ASCVD pathogenesis, with lifetime exposure as a critical determinant of risk.^5,6 Moreover, randomized controlled trials of statins and other studies of cholesterol-lowering show a reproducible relationship between the LDL-C level achieved and absolute risk (Figure 1).^7–24 We believe the totality of data constitutes a strong rationale for targeting LDL-C and establishing goals for lowering its levels. For these reasons, we believe that removing LDL-C goals is a fundamental flaw of the new guidelines.

The reason for the lack of data from randomized controlled trials demonstrating benefits of adding therapies to statins (when LDL-C is still high) or benefits of treating to specific goals is that no such trials have been performed. Even trials of nonpharmacologic means of lowering LDL-C, such as ileal bypass, which was used in the Program on the Surgical Control of the Hyperlipidemias trial,²⁰ provide independent evidence that lowering LDL-C reduces the risk of ASCVD (Figure 1).

In addition, trials of nonstatin drugs, such as the Coronary Drug Project,25 which tested niacin, also showed outcome benefits. On the other hand, studies such as the Atherothrombosis Intervention in Metabolic Syndrome With Low HDL/High Triglycerides: Impact on Global Health26 and Treatment of HDL to Reduce the Incidence of Vascular Events27 trials did not show additional risk reduction when niacin was added to statin therapy. However, the study designs arguably had flaws, including requirement of aggressive LDL-lowering with statins, with LDL-C levels below 70 to 80 mg/dL before randomization.

Therefore, these trials do not tell us what to do for a patient on maximal intensive therapy who has recurrent ASCVD events or who, like our patient, has an LDL-C level higher than previous targets.

For this patient, we would recommend adding a second medication to further lower his LDL-C, but discussing with him the absence of proven benefit in clinical trials and the risks of side effects. At present, lacking LDL-C goals in the new guidelines, we are keeping with the ATP III goals to help guide therapeutic choices and individualize patient management.

GROUP 2: PEOPLE WITH LDL-C ≥ 190

Advantages of the new guidelines

• They state that these patients should receive statins in the highest tolerated doses, which is universally accepted.

Limitations of the new guidelines

• The new guidelines mention only that one “may consider” adding a second agent if LDL-C remains above 190 mg/dL after maximum-dose therapy. Patients with familial hypercholesterolemia or other severe forms of hypercholesterolemia typically end up on multidrug therapy to further reduce LDL-C. The absence of randomized controlled trial data in this setting to show an additive value of second and third lipid-lowering agents does not mean these agents do not provide benefit.

GROUP 3: DIABETES, AGE 40–75, LDL-C 70–189, NO CLINICAL ASCVD

Advantages of the new guidelines

• They call for aggressive treatment of people with diabetes, a group at high risk that derives significant benefit from statin therapy, as shown in randomized controlled trials.

Limitations of the new guidelines

• Although high-intensity statin therapy is indicated for this group, we believe that, using the new risk calculator, some patients may receive overly aggressive treatment, thus increasing the possibility of statin side effects.
• The guidelines do not address patients younger than 40 or older than 75.
• Diabetic patients have a high residual risk of ASCVD events, even on statin therapy. Yet the guidelines ignore the potential benefits of more aggressive LDL-lowering or non-LDL secondary targets for therapy.

Case 2: How low is too low?

A 63-year-old white woman, a nonsmoker with recently diagnosed diabetes, is seen by her primary care physician. She has hypertension, for which she takes lisinopril 5 mg daily. Her fasting lipid values are:

• Total cholesterol 160 mg/dL
• HDL-C 64 mg/dL
• Triglycerides 100 mg/dL
• Calculated LDL-C 76 mg/dL.

Her systolic blood pressure is 129 mm Hg, and based on the new risk calculator, her 10-year risk of cardiovascular disease is 10.2%. According to the new guidelines, she should be started on high-intensity statin treatment (Table 1).

Although this is an acceptable initial course of action, it necessitates close vigilance, since it may actually drive her LDL-C level too low. Randomized controlled trials have typically used an LDL-C concentration of less than or equal to 25 mg/dL as the safety cutoff. With a typical LDL-C reduction of at least 50% on high-intensity statins, our patient’s expected LDL-C level will likely be in the low 30s. We believe this would be a good outcome, provided that she tolerates the medication without adverse effects. However, responses to statins vary from patient to patient.

High-intensity statin therapy may not be necessary to reduce risk adequately in all patients who have diabetes without preexisting vascular disease. The Collaborative Atorvastatin Diabetes Study¹² compared atorvastatin 10 mg vs placebo in people with type 2 diabetes, age 40 to 75, who had one or more cardiovascular risk factors but no signs or symptoms of preexisting ASCVD and who had only average or below-average cholesterol levels—precisely like this patient. The trial was terminated early because of a clear benefit (a 37% reduction in the composite end point of major adverse cardiovascular events) in the intervention group. For our patient, we believe an alternative and acceptable approach would be to begin moderate-intensity statin therapy (eg, with atorvastatin 10 mg) (Table 1).

Alternatively, in a patient with diabetes and previous atherosclerotic vascular disease or with a high 10-year risk and high LDL-C, limiting treatment to high-intensity statin therapy by itself may deny them the potential benefits of combination therapies and targeting to lower LDL-C levels or non-HDL-C secondary targets. Guidelines from the American Diabetes Association²⁸ and the American Association of Clinical Endocrinologists²⁹ continue to recommend an LDL-C goal of less than 70 mg/dL in patients at high risk, a non-HDL-C less than 100 mg/dL, an apolipoprotein B less than 80 mg/dL, and an LDL particle number less than 1,000 nmol/L.

GROUP 4: AGE 40–75, LDL-C 70–189, NO ASCVD, BUT 10-YEAR RISK ≥ 7.5%

Advantages of the new guidelines

• They may reduce ASCVD events for patients at higher risk.
• The risk calculator is easy to use and focuses on global risk, ie, all forms of ASCVD.
• The guidelines promote discussion of risks and benefits between patients and providers.

Limitations of the new guidelines

• The new risk calculator is controversial (see below).
• There is potential for overtreatment, particularly in older patients.
• There is potential for undertreatment, particularly in patients with an elevated LDL-C but whose 10-year risk is less than 7.5% because they are young.
• The guidelines do not address patients younger than 40 or older than 75.
• They do not take into account some traditional risk factors, such as family history, and nontraditional risk factors such as C-reactive protein as measured by ultrasensitive assays, lipoprotein(a), and apolipoprotein B.

Risk calculator controversy

The new risk calculator has aroused strong opinions on both sides of the aisle.

Shortly after the new guidelines were released, cardiologists Dr. Paul Ridker and Dr. Nancy Cook from Brigham and Women’s Hospital in Boston published analyses³⁰ showing that the new risk calculator, which was based on older data from several large cohorts such as the Atherosclerosis Risk in Communities study,³¹ the Cardiovascular Health Study,³² the Coronary Artery Risk Development in Young Adults study,³³ and the Framingham Heart Study,^34,35 was inaccurate in other cohorts. Specifically, in more-recent cohorts (the Women’s Health Study,³⁶ Physicians’ Health Study,³⁷ and Women’s Health Initiative³⁸), the new calculator overestimates the 10-year risk of ASCVD by 75% to 150%.³⁰ Using the new calculator would make approximately 30 million more Americans eligible for statin treatment. The concern is that patients at lower risk would be treated and exposed to potential side effects of statin therapy.

In addition, the risk calculator relies heavily on age and sex and does not include other factors such as triglyceride level, family history, C-reactive protein, or lipoprotein(a). Importantly, and somewhat ironically given the otherwise absolute adherence to randomized controlled trial data for guideline development, the risk calculator has never been verified in prospective studies to adequately show that using it reduces ASCVD events.

Based on the risk calculator, essentially any African American man in his early 60s with no other risk factors has a 10-year risk of ASCVD of 7.5% or higher and, according to the new guidelines, should receive at least moderate-intensity statin therapy.

For example, consider a 64-year-old African American man whose systolic blood pressure is 129 mm Hg, who does not smoke, does not have diabetes, and does not have hypertension, and whose total cholesterol level is 180 mg/dL, HDL-C 70 mg/dL, triglycerides 130 mg/dL, and calculated LDL-C 84 mg/dL. His calculated 10-year risk is, surprisingly, 7.5%.

Alternatively, his twin brother is a two-pack-per-day smoker with untreated hypertension and systolic blood pressure 150 mm Hg, with fasting total cholesterol 153 mg/dL, HDL-C 70 mg/dL, triglycerides 60 mg/dL, and LDL-C 71 mg/dL. His calculated 10-year risk is 10.5%, so according to the new guidelines, he too should receive high-intensity statin therapy. Yet this patient clearly needs better blood pressure control and smoking cessation as his primary risk-reduction efforts, not a statin. While assessing global risk is important, a shortcoming of the new guidelines is that they can inappropriately lead to treating the risk score, not individualizing the treatment to the patient. Because of the errors inherent in the risk calculator, some experts have called for a temporary halt on implementing the new guidelines until the risk calculator can be further validated. In November 2013, the American Heart Association and the American College of Cardiology reaffirmed their support of the new guidelines and recommended that they be implemented as planned. As of the time this manuscript goes to print, there are no plans to halt implementation of the new guidelines.

Case 4: Undertreating a primary prevention patient

A 25-year-old white man with no medical history has a total cholesterol level of 310 mg/dL, HDL-C 50 mg/dL, triglycerides 400 mg/dL, and calculated LDL-C 180 mg/dL. He does not smoke or have hypertension or diabetes but has a strong family history of premature coronary disease (his father died of myocardial infarction at age 42). His body mass index is 25 kg/m². Because he is less than 40 years old, the risk calculator does not apply to him.

If we assume he remains untreated and returns at age 40 with the same clinical factors and laboratory values, his calculated 10-year risk of an ASCVD event according to the new risk calculator will still be only 3.1%. Assuming his medical history remains unchanged as he continues to age, his 10-year risk would not reach 7.5% until he is 58. Would you feel comfortable waiting 33 years before starting statin therapy in this patient?

Waiting for dyslipidemic patients to reach middle age before starting LDL-C-lowering therapy is a failure of prevention. For practical reasons, there are no data from randomized controlled trials with hard outcomes in younger people. Nevertheless, a tenet of preventive cardiology is that cumulative exposure accelerates the “vascular age” ahead of the chronological age. This case illustrates why individualized recommendations guided by LDL-C goals as a target for therapy are needed. For this 25-year-old patient, we would recommend starting an intermediate- or high-potency statin.

Case 5: Rheumatoid arthritis

A 60-year-old postmenopausal white woman with severe rheumatoid arthritis presents for cholesterol evaluation. Her total cholesterol level is 235 mg/dL, HDL-C 50 mg/dL, and LDL-C 165 mg/dL. She does not smoke or have hypertension or diabetes. Her systolic blood pressure is 110 mm Hg. She has elevated C-reactive protein on an ultrasensitive assay and elevated lipoprotein(a).

Her calculated 10-year risk of ASCVD is 3.0%. Assuming her medical history remains the same, she would not reach a calculated 10-year risk of at least 7.5% until age 70. We suggest starting moderate- or high-dose statin therapy in this case, based on data (not from randomized controlled trials) showing an increased risk of ASCVD events in patients with rheumatologic disease, increased lipoprotein(a), and inflammatory markers like C-reactive protein. However, the current guidelines do not address this scenario, other than to suggest that clinician consideration can be given to other risk markers such as these, and that these findings should be discussed in detail with the patient. The Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin trial¹⁶ showed a dramatic ASCVD risk reduction in just such patients (Figure 1).

APPLAUSE—AND RESERVATIONS

The newest guidelines for treating high blood cholesterol represent a monumental shift away from using target levels of LDL-C and non-HDL-C and toward a focus on statin intensity for patients in the four highest-risk groups.

We applaud the expert panel for its idealistic approach of using only data from randomized controlled trials, for placing more emphasis on higher-intensity statin treatment, for including stroke in the new definition of ASCVD, and for focusing more attention on treating diabetic patients more aggressively. Simplifying the guidelines is a noble goal. Emphasizing moderate-to-high-intensity statin therapy in patients at moderate-to-high risk should have substantial long-term public health benefits.

However, as we have shown in the case examples, there are significant limitations, and some patients can end up being overtreated, while others may be undertreated.

Guidelines need to be crafted by looking at all the evidence, including the pathophysiology of the disease process, not just data from randomized controlled trials. It is difficult to implement a guideline that on one hand used randomized controlled trials exclusively for recommendations, but on the other hand used an untested risk calculator to guide therapy. Randomized controlled trials are not available for every scenario.

Further, absence of randomized controlled trial data in a given scenario should not be interpreted as evidence of lack of benefit. An example of this is a primary-prevention patient under age 40 with elevated LDL-C below the 190 mg/dL cutoff who otherwise is healthy and without risk factors (eg, Case 4). By disregarding all evidence that is not from randomized controlled trials, the expert panel fails to account for the extensive pathophysiology of ASCVD, which often begins at a young age and takes decades to develop.^5,6,39 An entire generation of patients who have not reached the age of inclusion in most randomized controlled trials with hard outcomes is excluded (unless the LDL-C level is very high), potentially setting back decades of progress in the field of prevention. Prevention only works if started. With childhood and young adult obesity sharply rising, we should not fail to address the under-40-year-old patient population in our guidelines.

Guidelines are designed to be expert opinion, not to dictate practice. Focusing on the individual patient instead of the general population at risk, the expert panel appropriately emphasizes the “importance of clinician judgment, weighing potential benefits, adverse effects, drug-drug interactions and patient preferences.” However, by excluding all data that do not come from randomized controlled trials, the panel neglects a very large base of knowledge and leaves many clinicians without as much expert opinion as we had hoped for.

LDL-C goals are important: they provide a scorecard to help the patient with lifestyle and dietary changes. They provide the health care provider guidance in making treatment decisions and focusing on treatment of a single patient, not a population. Moreover, if a patient has difficulty taking standard doses of statins because of side effects, the absence of LDL-C goals makes decision-making nearly impossible. We hope physicians will rely on LDL-C goals in such situations, falling back on the ATP III recommendations, although many patients may simply go untreated until they present with ASCVD or until they “age in” to a higher risk category.

We suggest caution in strict adherence to the new guidelines and instead urge physicians to consider a hybrid of the old guidelines (using the ATP III LDL-C goals) and the new ones (emphasizing global risk assessment and high-intensity statin treatment).

On November 12, 2013, a joint task force for the American College of Cardiology and American Heart Association released new guidelines for treating high blood cholesterol to reduce the risk of atherosclerotic cardiovascular disease (ASCVD) in adults.¹

This document arrives after several years of intense deliberation, 12 years after the third Adult Treatment Panel (ATP III) guidelines,² and 8 years after an ATP III update recommending that low-density lipoprotein cholesterol (LDL-C) levels be lowered aggressively (to less than 70 mg/dL) as an option in patients at high risk.³ It represents a major shift in the approach to and management of high blood cholesterol and has sparked considerable controversy.

In the following commentary, we summarize the new guidelines and the philosophy employed by the task force in generating them. We will also examine some advantages and what we believe to be several shortcomings of the new guidelines. These latter points are illustrated through case examples.

IN RANDOMIZED CONTROLLED TRIALS WE TRUST

In collaboration with the National Heart, Lung, and Blood Institute of the National Institutes of Health, the American College of Cardiology and American Heart Association formed an expert panel task force in 2008.

The task force elected to use only evidence from randomized controlled trials, systematic reviews, and meta-analyses of randomized controlled trials (and only predefined outcomes of the trials, not post hoc analyses) in formulating its recommendations, with the goal of providing the strongest possible evidence.

The authors state that “By using [randomized controlled trial] data to identify those most likely to benefit [emphasis in original] from cholesterol-lowering statin therapy, the recommendations will be of value to primary care clinicians as well as specialists concerned with ASCVD prevention. Importantly, the recommendations were designed to be easy to use in the clinical setting, facilitating the implementation of a strategy of risk assessment and treatment focused on the prevention of ASCVD.”³ They also state the guidelines are meant to “inform clinical judgment, not replace it” and that clinician judgment in addition to discussion with patients remains vital.

During the deliberations, the National Heart, Lung, and Blood Institute removed itself from participating, stating its mission no longer included drafting new guidelines. Additionally, other initial members of the task force removed themselves because of disagreement and concerns about the direction of the new guidelines.

These guidelines, and their accompanying new cardiovascular risk calculator,⁴ were released without a preliminary period to allow for open discussion, comment, and critique by physicians outside the panel. No attempt was made to harmonize the guidelines with previous versions (eg, ATP III) or with current international guidelines.

WHAT’S NEW IN THE GUIDELINES?

The following are the major changes in the new guidelines for treating high blood cholesterol:

• Treatment goals for LDL-C and non-high-density lipoprotein cholesterol (non-HDL-C) are no longer recommended.
• High-intensity and moderate-intensity statin treatment is emphasized, and low-intensity statin therapy is nearly eliminated.
• “ASCVD” now includes stroke in addition to coronary heart disease and peripheral arterial disease.
• Four groups are targeted for treatment (see below).
• Nonstatin therapies have been markedly de-emphasized.
• No guidelines are provided for treating high triglyceride levels.

The new guidelines emphasize lifestyle modification as the foundation for reducing risk, regardless of cholesterol therapy. No recommendations are given for patients with New York Heart Association class II, III, or IV heart failure or for hemodialysis patients, because there were insufficient data from randomized controlled trials to support recommendations. Similarly, the guidelines apply only to people between the ages of 40 and 75 (risk calculator ages 40–79), because the authors believed there was not enough evidence from randomized controlled trials to allow development of guidelines outside of this age range.

FOUR MAJOR STATIN TREATMENT GROUPS

The new guidelines specify four groups that merit intensive or moderately intensive statin therapy (Table 1)¹:

• People with clinical ASCVD
• People with LDL-C levels of 190 mg/dL or higher
• People with diabetes, age 40 to 75
• People without diabetes, age 40 to 75, with LDL-C levels 70–189 mg/dL, and a 10-year ASCVD risk of 7.5% or higher as determined by the new risk calculator⁴ (which also calculates the lifetime risk of ASCVD).

Below, we will address each of these four groups and provide case scenarios to consider. In general, our major disagreements with the new recommendations pertain to the first and fourth categories.

GROUP 1: PEOPLE WITH CLINICAL ASCVD

Advantages of the new guidelines

• They appropriately recommend statins in the highest tolerated doses as first-line treatment for this group at high risk.
• They designate all patients with ASCVD, including those with coronary, peripheral, and cerebrovascular disease, as a high-risk group.
• Without target LDL-C levels, treatment is simpler than before, requiring less monitoring of lipid levels. (This can also be seen as a limitation, as we discuss below.)

Limitations of the new guidelines

• They make follow-up LDL-C levels irrelevant, seeming to assume that there is no gradation in residual risk and, thus, no need to tailor therapy to the individual.
• Patients no longer have a goal to strive for or a way to monitor their progress.
• The guidelines ignore the pathophysiology of coronary artery disease and evidence of residual risk in patients on both moderate-intensity and high-intensity statin therapy.
• They also ignore the potential benefits of treating to lower LDL-C or non-HDL-C goals, thus eliminating consideration of multidrug therapy. They do not address patients with recurrent cardiovascular events already on maximal tolerated statin doses.
• They undermine the potential development and use of new therapies for dysplipidemia in patients with ASCVD.

Case 1: Is LDL-C 110 mg/dL low enough?

A 52-year-old African American man presents with newly discovered moderate coronary artery disease that is not severe enough to warrant stenting. He has no history of hypertension, diabetes mellitus, or smoking. His systolic blood pressure is 130 mm Hg, and his body mass index is 26 kg/m². He exercises regularly and follows a low-cholesterol diet. He has the following fasting lipid values:

• Total cholesterol 290 mg/dL
• HDL-C 50 mg/dL
• Triglycerides 250 mg/dL
• Calculated LDL-C 190 mg/dL.

Two months later, after beginning atorvastatin 80 mg daily, meeting with a nutritionist, and redoubling his dietary efforts, his fasting lipid concentrations are:

• Total cholesterol 180 mg/dL
• HDL-C 55 mg/dL
• Triglycerides 75 mg/dL
• Calculated LDL-C 110 mg/dL.

Comment: Lack of LDL-C goals is a flaw

The new guidelines call for patients with known ASCVD, such as this patient, to receive intensive statin therapy—which he got.

However, once a patient is on therapy, the new guidelines do not encourage repeating the lipid panel other than to assess compliance. With intensive therapy, we expect a reduction in LDL-C of at least 50% (Table 1), but patient-to-patient differences in response to medications are common, and without repeat testing we would have no way of gauging this patient’s residual risk.

Further, the new guidelines emphasize the lack of hard outcome data supporting the addition of another lipid-lowering drug to a statin, although they do indicate that one can consider it. In a patient at high risk, such as this one, would you be comfortable with an LDL-C value of 110 mg/dL on maximum statin therapy? Would you consider adding a nonstatin drug?

A preponderance of data shows that LDL plays a causal role in ASCVD development and adverse events. Genetic data show that the LDL particle and the LDL receptor pathway are mechanistically linked to ASCVD pathogenesis, with lifetime exposure as a critical determinant of risk.^5,6 Moreover, randomized controlled trials of statins and other studies of cholesterol-lowering show a reproducible relationship between the LDL-C level achieved and absolute risk (Figure 1).^7–24 We believe the totality of data constitutes a strong rationale for targeting LDL-C and establishing goals for lowering its levels. For these reasons, we believe that removing LDL-C goals is a fundamental flaw of the new guidelines.

The reason for the lack of data from randomized controlled trials demonstrating benefits of adding therapies to statins (when LDL-C is still high) or benefits of treating to specific goals is that no such trials have been performed. Even trials of nonpharmacologic means of lowering LDL-C, such as ileal bypass, which was used in the Program on the Surgical Control of the Hyperlipidemias trial,²⁰ provide independent evidence that lowering LDL-C reduces the risk of ASCVD (Figure 1).

In addition, trials of nonstatin drugs, such as the Coronary Drug Project,25 which tested niacin, also showed outcome benefits. On the other hand, studies such as the Atherothrombosis Intervention in Metabolic Syndrome With Low HDL/High Triglycerides: Impact on Global Health26 and Treatment of HDL to Reduce the Incidence of Vascular Events27 trials did not show additional risk reduction when niacin was added to statin therapy. However, the study designs arguably had flaws, including requirement of aggressive LDL-lowering with statins, with LDL-C levels below 70 to 80 mg/dL before randomization.

Therefore, these trials do not tell us what to do for a patient on maximal intensive therapy who has recurrent ASCVD events or who, like our patient, has an LDL-C level higher than previous targets.

For this patient, we would recommend adding a second medication to further lower his LDL-C, but discussing with him the absence of proven benefit in clinical trials and the risks of side effects. At present, lacking LDL-C goals in the new guidelines, we are keeping with the ATP III goals to help guide therapeutic choices and individualize patient management.

GROUP 2: PEOPLE WITH LDL-C ≥ 190

Advantages of the new guidelines

• They state that these patients should receive statins in the highest tolerated doses, which is universally accepted.

Limitations of the new guidelines

• The new guidelines mention only that one “may consider” adding a second agent if LDL-C remains above 190 mg/dL after maximum-dose therapy. Patients with familial hypercholesterolemia or other severe forms of hypercholesterolemia typically end up on multidrug therapy to further reduce LDL-C. The absence of randomized controlled trial data in this setting to show an additive value of second and third lipid-lowering agents does not mean these agents do not provide benefit.

GROUP 3: DIABETES, AGE 40–75, LDL-C 70–189, NO CLINICAL ASCVD

Advantages of the new guidelines

• They call for aggressive treatment of people with diabetes, a group at high risk that derives significant benefit from statin therapy, as shown in randomized controlled trials.

Limitations of the new guidelines

• Although high-intensity statin therapy is indicated for this group, we believe that, using the new risk calculator, some patients may receive overly aggressive treatment, thus increasing the possibility of statin side effects.
• The guidelines do not address patients younger than 40 or older than 75.
• Diabetic patients have a high residual risk of ASCVD events, even on statin therapy. Yet the guidelines ignore the potential benefits of more aggressive LDL-lowering or non-LDL secondary targets for therapy.

Case 2: How low is too low?

A 63-year-old white woman, a nonsmoker with recently diagnosed diabetes, is seen by her primary care physician. She has hypertension, for which she takes lisinopril 5 mg daily. Her fasting lipid values are:

• Total cholesterol 160 mg/dL
• HDL-C 64 mg/dL
• Triglycerides 100 mg/dL
• Calculated LDL-C 76 mg/dL.

Her systolic blood pressure is 129 mm Hg, and based on the new risk calculator, her 10-year risk of cardiovascular disease is 10.2%. According to the new guidelines, she should be started on high-intensity statin treatment (Table 1).

Although this is an acceptable initial course of action, it necessitates close vigilance, since it may actually drive her LDL-C level too low. Randomized controlled trials have typically used an LDL-C concentration of less than or equal to 25 mg/dL as the safety cutoff. With a typical LDL-C reduction of at least 50% on high-intensity statins, our patient’s expected LDL-C level will likely be in the low 30s. We believe this would be a good outcome, provided that she tolerates the medication without adverse effects. However, responses to statins vary from patient to patient.

High-intensity statin therapy may not be necessary to reduce risk adequately in all patients who have diabetes without preexisting vascular disease. The Collaborative Atorvastatin Diabetes Study¹² compared atorvastatin 10 mg vs placebo in people with type 2 diabetes, age 40 to 75, who had one or more cardiovascular risk factors but no signs or symptoms of preexisting ASCVD and who had only average or below-average cholesterol levels—precisely like this patient. The trial was terminated early because of a clear benefit (a 37% reduction in the composite end point of major adverse cardiovascular events) in the intervention group. For our patient, we believe an alternative and acceptable approach would be to begin moderate-intensity statin therapy (eg, with atorvastatin 10 mg) (Table 1).

Alternatively, in a patient with diabetes and previous atherosclerotic vascular disease or with a high 10-year risk and high LDL-C, limiting treatment to high-intensity statin therapy by itself may deny them the potential benefits of combination therapies and targeting to lower LDL-C levels or non-HDL-C secondary targets. Guidelines from the American Diabetes Association²⁸ and the American Association of Clinical Endocrinologists²⁹ continue to recommend an LDL-C goal of less than 70 mg/dL in patients at high risk, a non-HDL-C less than 100 mg/dL, an apolipoprotein B less than 80 mg/dL, and an LDL particle number less than 1,000 nmol/L.

GROUP 4: AGE 40–75, LDL-C 70–189, NO ASCVD, BUT 10-YEAR RISK ≥ 7.5%

Advantages of the new guidelines

• They may reduce ASCVD events for patients at higher risk.
• The risk calculator is easy to use and focuses on global risk, ie, all forms of ASCVD.
• The guidelines promote discussion of risks and benefits between patients and providers.

Limitations of the new guidelines

• The new risk calculator is controversial (see below).
• There is potential for overtreatment, particularly in older patients.
• There is potential for undertreatment, particularly in patients with an elevated LDL-C but whose 10-year risk is less than 7.5% because they are young.
• The guidelines do not address patients younger than 40 or older than 75.
• They do not take into account some traditional risk factors, such as family history, and nontraditional risk factors such as C-reactive protein as measured by ultrasensitive assays, lipoprotein(a), and apolipoprotein B.

Risk calculator controversy

The new risk calculator has aroused strong opinions on both sides of the aisle.

Shortly after the new guidelines were released, cardiologists Dr. Paul Ridker and Dr. Nancy Cook from Brigham and Women’s Hospital in Boston published analyses³⁰ showing that the new risk calculator, which was based on older data from several large cohorts such as the Atherosclerosis Risk in Communities study,³¹ the Cardiovascular Health Study,³² the Coronary Artery Risk Development in Young Adults study,³³ and the Framingham Heart Study,^34,35 was inaccurate in other cohorts. Specifically, in more-recent cohorts (the Women’s Health Study,³⁶ Physicians’ Health Study,³⁷ and Women’s Health Initiative³⁸), the new calculator overestimates the 10-year risk of ASCVD by 75% to 150%.³⁰ Using the new calculator would make approximately 30 million more Americans eligible for statin treatment. The concern is that patients at lower risk would be treated and exposed to potential side effects of statin therapy.

In addition, the risk calculator relies heavily on age and sex and does not include other factors such as triglyceride level, family history, C-reactive protein, or lipoprotein(a). Importantly, and somewhat ironically given the otherwise absolute adherence to randomized controlled trial data for guideline development, the risk calculator has never been verified in prospective studies to adequately show that using it reduces ASCVD events.

Based on the risk calculator, essentially any African American man in his early 60s with no other risk factors has a 10-year risk of ASCVD of 7.5% or higher and, according to the new guidelines, should receive at least moderate-intensity statin therapy.

For example, consider a 64-year-old African American man whose systolic blood pressure is 129 mm Hg, who does not smoke, does not have diabetes, and does not have hypertension, and whose total cholesterol level is 180 mg/dL, HDL-C 70 mg/dL, triglycerides 130 mg/dL, and calculated LDL-C 84 mg/dL. His calculated 10-year risk is, surprisingly, 7.5%.

Alternatively, his twin brother is a two-pack-per-day smoker with untreated hypertension and systolic blood pressure 150 mm Hg, with fasting total cholesterol 153 mg/dL, HDL-C 70 mg/dL, triglycerides 60 mg/dL, and LDL-C 71 mg/dL. His calculated 10-year risk is 10.5%, so according to the new guidelines, he too should receive high-intensity statin therapy. Yet this patient clearly needs better blood pressure control and smoking cessation as his primary risk-reduction efforts, not a statin. While assessing global risk is important, a shortcoming of the new guidelines is that they can inappropriately lead to treating the risk score, not individualizing the treatment to the patient. Because of the errors inherent in the risk calculator, some experts have called for a temporary halt on implementing the new guidelines until the risk calculator can be further validated. In November 2013, the American Heart Association and the American College of Cardiology reaffirmed their support of the new guidelines and recommended that they be implemented as planned. As of the time this manuscript goes to print, there are no plans to halt implementation of the new guidelines.

Case 4: Undertreating a primary prevention patient

A 25-year-old white man with no medical history has a total cholesterol level of 310 mg/dL, HDL-C 50 mg/dL, triglycerides 400 mg/dL, and calculated LDL-C 180 mg/dL. He does not smoke or have hypertension or diabetes but has a strong family history of premature coronary disease (his father died of myocardial infarction at age 42). His body mass index is 25 kg/m². Because he is less than 40 years old, the risk calculator does not apply to him.

If we assume he remains untreated and returns at age 40 with the same clinical factors and laboratory values, his calculated 10-year risk of an ASCVD event according to the new risk calculator will still be only 3.1%. Assuming his medical history remains unchanged as he continues to age, his 10-year risk would not reach 7.5% until he is 58. Would you feel comfortable waiting 33 years before starting statin therapy in this patient?

Waiting for dyslipidemic patients to reach middle age before starting LDL-C-lowering therapy is a failure of prevention. For practical reasons, there are no data from randomized controlled trials with hard outcomes in younger people. Nevertheless, a tenet of preventive cardiology is that cumulative exposure accelerates the “vascular age” ahead of the chronological age. This case illustrates why individualized recommendations guided by LDL-C goals as a target for therapy are needed. For this 25-year-old patient, we would recommend starting an intermediate- or high-potency statin.

Case 5: Rheumatoid arthritis

A 60-year-old postmenopausal white woman with severe rheumatoid arthritis presents for cholesterol evaluation. Her total cholesterol level is 235 mg/dL, HDL-C 50 mg/dL, and LDL-C 165 mg/dL. She does not smoke or have hypertension or diabetes. Her systolic blood pressure is 110 mm Hg. She has elevated C-reactive protein on an ultrasensitive assay and elevated lipoprotein(a).

Her calculated 10-year risk of ASCVD is 3.0%. Assuming her medical history remains the same, she would not reach a calculated 10-year risk of at least 7.5% until age 70. We suggest starting moderate- or high-dose statin therapy in this case, based on data (not from randomized controlled trials) showing an increased risk of ASCVD events in patients with rheumatologic disease, increased lipoprotein(a), and inflammatory markers like C-reactive protein. However, the current guidelines do not address this scenario, other than to suggest that clinician consideration can be given to other risk markers such as these, and that these findings should be discussed in detail with the patient. The Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin trial¹⁶ showed a dramatic ASCVD risk reduction in just such patients (Figure 1).

APPLAUSE—AND RESERVATIONS

The newest guidelines for treating high blood cholesterol represent a monumental shift away from using target levels of LDL-C and non-HDL-C and toward a focus on statin intensity for patients in the four highest-risk groups.

We applaud the expert panel for its idealistic approach of using only data from randomized controlled trials, for placing more emphasis on higher-intensity statin treatment, for including stroke in the new definition of ASCVD, and for focusing more attention on treating diabetic patients more aggressively. Simplifying the guidelines is a noble goal. Emphasizing moderate-to-high-intensity statin therapy in patients at moderate-to-high risk should have substantial long-term public health benefits.

However, as we have shown in the case examples, there are significant limitations, and some patients can end up being overtreated, while others may be undertreated.

Guidelines need to be crafted by looking at all the evidence, including the pathophysiology of the disease process, not just data from randomized controlled trials. It is difficult to implement a guideline that on one hand used randomized controlled trials exclusively for recommendations, but on the other hand used an untested risk calculator to guide therapy. Randomized controlled trials are not available for every scenario.

Further, absence of randomized controlled trial data in a given scenario should not be interpreted as evidence of lack of benefit. An example of this is a primary-prevention patient under age 40 with elevated LDL-C below the 190 mg/dL cutoff who otherwise is healthy and without risk factors (eg, Case 4). By disregarding all evidence that is not from randomized controlled trials, the expert panel fails to account for the extensive pathophysiology of ASCVD, which often begins at a young age and takes decades to develop.^5,6,39 An entire generation of patients who have not reached the age of inclusion in most randomized controlled trials with hard outcomes is excluded (unless the LDL-C level is very high), potentially setting back decades of progress in the field of prevention. Prevention only works if started. With childhood and young adult obesity sharply rising, we should not fail to address the under-40-year-old patient population in our guidelines.

Guidelines are designed to be expert opinion, not to dictate practice. Focusing on the individual patient instead of the general population at risk, the expert panel appropriately emphasizes the “importance of clinician judgment, weighing potential benefits, adverse effects, drug-drug interactions and patient preferences.” However, by excluding all data that do not come from randomized controlled trials, the panel neglects a very large base of knowledge and leaves many clinicians without as much expert opinion as we had hoped for.

LDL-C goals are important: they provide a scorecard to help the patient with lifestyle and dietary changes. They provide the health care provider guidance in making treatment decisions and focusing on treatment of a single patient, not a population. Moreover, if a patient has difficulty taking standard doses of statins because of side effects, the absence of LDL-C goals makes decision-making nearly impossible. We hope physicians will rely on LDL-C goals in such situations, falling back on the ATP III recommendations, although many patients may simply go untreated until they present with ASCVD or until they “age in” to a higher risk category.

We suggest caution in strict adherence to the new guidelines and instead urge physicians to consider a hybrid of the old guidelines (using the ATP III LDL-C goals) and the new ones (emphasizing global risk assessment and high-intensity statin treatment).

In Reply: As Dr. Thacker notes, women are underrepresented in statin clinical trials. This, in addition to the fact that the metaanalyses reviewed did not generally stratify results by sex, makes a detailed discussion of sex-based differences on diabetes incidence and comparative outcomes difficult.

In terms of outcomes, some metaanalyses have found similar reductions of cardiovascular events with statin treatment in men and women, particularly in secondary-prevention populations.^1–3 Even though the cited report from Gutierrez et al⁴ may not have been as inclusive as some other studies, it also demonstrated similar reductions in myocardial infarction, need for intervention, and coronary mortality rates compared with men. The lack of significant reduction in rates of cerebrovascular accidents and all-cause mortality in this study may be a function of the low percentage of women in the analysis (20.6%), the low number of events, and the lack of power. However, the results did trend in a positive direction.

It is true that outcome benefits are harder to demonstrate in primary-prevention populations. However, a meta-analysis by Brugts et al⁵ in 2009 examined 10 placebocontrolled statin trials, including at least 80% of individuals without cardiovascular disease or whose data were reported from a sole primary prevention group. Thirty-four percent of the participants were women. Overall, there was a 12% reduction in mortality, 30% reduction in coronary events, and 19% reduction in cerebrovascular events. Although sex-specific analysis did not show significant reductions in women alone, the directional trends were similar to those in men, and subgroup analysis revealed no heterogeneity in treatment effect by sex, age, or diabetes status.

The meta-analysis from the Cholesterol Treatment Trialists cited in this review included 27 controlled trials and stratified patients by estimated 5-year major vascular event risk⁶; 29% of the patients were women. As expected, annual event rates increased with increasing estimate of risk. Rates of major vascular and major coronary events were reduced by 21% and 30%, respectively. Similar significant proportional reductions were noted in all risk groups, including the lowest two risk groups (< 5% and 5 to < 10%). Although analysis was not stratified by sex, there was a proportionately higher percentage of women (54%) represented in the lowest-risk group, which had a similar relative risk reduction. In the primary-prevention trial JUPITER⁷ in patients with elevated C-reactive protein and low low-density lipoprotein cholesterol levels, rosuvastatin significantly reduced the primary composite end point in women (38% of the study group) by 46%, which was similar to that in the men.^7,8 In the same paper, an additional meta-analysis of exclusively primary prevention trials reported a significant 37% reduction in cardiovascular events.

As for comparable diabetes incidence on statins, it is not accurate to imply that women have a higher risk of developing diabetes than men based only on the Women’s Health Initiative observational analysis—an all-female study with no male comparison arm, without randomization to statins, and in which only 7% of participants at entry were taking the drug in question.

The use of statins in low-risk individuals and in women in particular does remain controversial, partially because of the lack of controlled data and sufficiently powered studies with women. It was not my intent to “promulgate a fiction” that statins should be used in primary prevention in all women, but rather to recommend the use of statins appropriately in at-risk patients after weighing the treatment risks. All therapies, including statin therapy, should be directed toward those who would have the best benefit-risk ratio. To lump together all primary-prevention women, however, is overly simplistic and may result in denying therapy to a patient who may benefit from the intervention. In women as in men, the available data (although imperfect) support statin use with an acceptable risk profile in those at moderate to high risk of subsequent cardiovascular events. Some of these patients would be in the primary prevention classification. Every decision to treat needs to factor in the patient’s overall cardiovascular risk, the likelihood of adverse effects including diabetes, and the patient’s sex.

In Reply: As Dr. Thacker notes, women are underrepresented in statin clinical trials. This, in addition to the fact that the metaanalyses reviewed did not generally stratify results by sex, makes a detailed discussion of sex-based differences on diabetes incidence and comparative outcomes difficult.

In terms of outcomes, some metaanalyses have found similar reductions of cardiovascular events with statin treatment in men and women, particularly in secondary-prevention populations.^1–3 Even though the cited report from Gutierrez et al⁴ may not have been as inclusive as some other studies, it also demonstrated similar reductions in myocardial infarction, need for intervention, and coronary mortality rates compared with men. The lack of significant reduction in rates of cerebrovascular accidents and all-cause mortality in this study may be a function of the low percentage of women in the analysis (20.6%), the low number of events, and the lack of power. However, the results did trend in a positive direction.

It is true that outcome benefits are harder to demonstrate in primary-prevention populations. However, a meta-analysis by Brugts et al⁵ in 2009 examined 10 placebocontrolled statin trials, including at least 80% of individuals without cardiovascular disease or whose data were reported from a sole primary prevention group. Thirty-four percent of the participants were women. Overall, there was a 12% reduction in mortality, 30% reduction in coronary events, and 19% reduction in cerebrovascular events. Although sex-specific analysis did not show significant reductions in women alone, the directional trends were similar to those in men, and subgroup analysis revealed no heterogeneity in treatment effect by sex, age, or diabetes status.

The meta-analysis from the Cholesterol Treatment Trialists cited in this review included 27 controlled trials and stratified patients by estimated 5-year major vascular event risk⁶; 29% of the patients were women. As expected, annual event rates increased with increasing estimate of risk. Rates of major vascular and major coronary events were reduced by 21% and 30%, respectively. Similar significant proportional reductions were noted in all risk groups, including the lowest two risk groups (< 5% and 5 to < 10%). Although analysis was not stratified by sex, there was a proportionately higher percentage of women (54%) represented in the lowest-risk group, which had a similar relative risk reduction. In the primary-prevention trial JUPITER⁷ in patients with elevated C-reactive protein and low low-density lipoprotein cholesterol levels, rosuvastatin significantly reduced the primary composite end point in women (38% of the study group) by 46%, which was similar to that in the men.^7,8 In the same paper, an additional meta-analysis of exclusively primary prevention trials reported a significant 37% reduction in cardiovascular events.

As for comparable diabetes incidence on statins, it is not accurate to imply that women have a higher risk of developing diabetes than men based only on the Women’s Health Initiative observational analysis—an all-female study with no male comparison arm, without randomization to statins, and in which only 7% of participants at entry were taking the drug in question.

The use of statins in low-risk individuals and in women in particular does remain controversial, partially because of the lack of controlled data and sufficiently powered studies with women. It was not my intent to “promulgate a fiction” that statins should be used in primary prevention in all women, but rather to recommend the use of statins appropriately in at-risk patients after weighing the treatment risks. All therapies, including statin therapy, should be directed toward those who would have the best benefit-risk ratio. To lump together all primary-prevention women, however, is overly simplistic and may result in denying therapy to a patient who may benefit from the intervention. In women as in men, the available data (although imperfect) support statin use with an acceptable risk profile in those at moderate to high risk of subsequent cardiovascular events. Some of these patients would be in the primary prevention classification. Every decision to treat needs to factor in the patient’s overall cardiovascular risk, the likelihood of adverse effects including diabetes, and the patient’s sex.

In Reply: As Dr. Thacker notes, women are underrepresented in statin clinical trials. This, in addition to the fact that the metaanalyses reviewed did not generally stratify results by sex, makes a detailed discussion of sex-based differences on diabetes incidence and comparative outcomes difficult.

In terms of outcomes, some metaanalyses have found similar reductions of cardiovascular events with statin treatment in men and women, particularly in secondary-prevention populations.^1–3 Even though the cited report from Gutierrez et al⁴ may not have been as inclusive as some other studies, it also demonstrated similar reductions in myocardial infarction, need for intervention, and coronary mortality rates compared with men. The lack of significant reduction in rates of cerebrovascular accidents and all-cause mortality in this study may be a function of the low percentage of women in the analysis (20.6%), the low number of events, and the lack of power. However, the results did trend in a positive direction.

It is true that outcome benefits are harder to demonstrate in primary-prevention populations. However, a meta-analysis by Brugts et al⁵ in 2009 examined 10 placebocontrolled statin trials, including at least 80% of individuals without cardiovascular disease or whose data were reported from a sole primary prevention group. Thirty-four percent of the participants were women. Overall, there was a 12% reduction in mortality, 30% reduction in coronary events, and 19% reduction in cerebrovascular events. Although sex-specific analysis did not show significant reductions in women alone, the directional trends were similar to those in men, and subgroup analysis revealed no heterogeneity in treatment effect by sex, age, or diabetes status.

The meta-analysis from the Cholesterol Treatment Trialists cited in this review included 27 controlled trials and stratified patients by estimated 5-year major vascular event risk⁶; 29% of the patients were women. As expected, annual event rates increased with increasing estimate of risk. Rates of major vascular and major coronary events were reduced by 21% and 30%, respectively. Similar significant proportional reductions were noted in all risk groups, including the lowest two risk groups (< 5% and 5 to < 10%). Although analysis was not stratified by sex, there was a proportionately higher percentage of women (54%) represented in the lowest-risk group, which had a similar relative risk reduction. In the primary-prevention trial JUPITER⁷ in patients with elevated C-reactive protein and low low-density lipoprotein cholesterol levels, rosuvastatin significantly reduced the primary composite end point in women (38% of the study group) by 46%, which was similar to that in the men.^7,8 In the same paper, an additional meta-analysis of exclusively primary prevention trials reported a significant 37% reduction in cardiovascular events.

As for comparable diabetes incidence on statins, it is not accurate to imply that women have a higher risk of developing diabetes than men based only on the Women’s Health Initiative observational analysis—an all-female study with no male comparison arm, without randomization to statins, and in which only 7% of participants at entry were taking the drug in question.

The use of statins in low-risk individuals and in women in particular does remain controversial, partially because of the lack of controlled data and sufficiently powered studies with women. It was not my intent to “promulgate a fiction” that statins should be used in primary prevention in all women, but rather to recommend the use of statins appropriately in at-risk patients after weighing the treatment risks. All therapies, including statin therapy, should be directed toward those who would have the best benefit-risk ratio. To lump together all primary-prevention women, however, is overly simplistic and may result in denying therapy to a patient who may benefit from the intervention. In women as in men, the available data (although imperfect) support statin use with an acceptable risk profile in those at moderate to high risk of subsequent cardiovascular events. Some of these patients would be in the primary prevention classification. Every decision to treat needs to factor in the patient’s overall cardiovascular risk, the likelihood of adverse effects including diabetes, and the patient’s sex.