Pleiotropic effects of statins in atherosclerotic disease

Abstract

Evaluation of: Puato M, Faggin E, Rattazzi M et al. Atorvastatin reduces macrophage accumulation in atherosclerotic plaques: a comparison of a nonstatin-based regimen in patients undergoing carotid endarterectomy. Stroke 41(6), 1163–1168 (2010).

Statins have an established role in the treatment of hypercholesterolemia and the prophylactic treatment of patients with atherosclerotic disease, and have been found to prevent secondary cardiovascular events and thereby reduce morbidity and mortality. Nevertheless, the pathophysiologic effect of statins on inflammatory responses and local atherosclerotic plaque morphology in humans remains a matter of debate. In particular the question is unanswered whether statin-induced alterations in plaque composition can be ascribed to LDL lowering or an anti-inflammatory pleiotropic effect. We will discuss the results of a recent study by Puato et al. concerning the effect of two different atorvastatin dosages and a nonstatin lipid-lowering drug on atherosclerotic plaque morphology.

Keywords::

• cardiovascular disease
• inflammation
• plaque composition
• statins

Hyperlipidemia is a well-recognized risk factor for cardiovascular disease and results in a consistent increase in plaque atheroma. Treatment with 3-hydroxy-3-methyl-glutaryl-CoA reductase inhibitors (statins) results in reduced cardiovascular mortality and morbidity [1]. Over the last 15 years the outcome of randomized trials has resulted in a consistent increase in the prescription and use of statins. Even in healthy persons with normal LDL levels, but with elevated high-sensitivity C-reactive protein (CRP) levels, statins reduced the incidence of cardiovascular events [2]. In addition to the lipid-lowering effects, statin treatment results in pleiotropic effects by modulating inflammatory responses that are involved in the initiation and progression of atherosclerotic disease. Human studies revealed that cytokine and CRP levels are lower in statin-treated patients compared with a placebo-treated patient group [3–5]. Experimental in vitro studies further support the concept that statins have anti-inflammatory properties [6,7].

In addition, multiple animal studies have shown that statin treatment results in an altered plaque phenotype with a less inflammatory stable plaque phenotype. For instance, in nonhuman primates it has been shown that statins reduce plaque inflammation independently of the effects on serum cholesterol [8,9].

The number of human studies with the objective to explore the pleiotropic effects of statins on atherosclerotic plaques is limited; it would be unethical to consider a study design including a placebo group when the beneficial effect of the treatment is evident. In a previous study, a small group of patients eligible for carotid surgery was randomized to receive either pravastatin or a placebo [10]. The researchers found more features of stable plaques in the pravastatin-treated group when the carotid lesion was studied following surgery. Comparisons of the plaque-stabilizing effects of low versus high dosages of statins have also been made. It has been shown that more aggressive atorvastatin treatment enhances plaque echogenicity when patients are treated with high dosages for 1 year and studied with carotid ultrasound [11]. Using serial MRI examinations it was shown that rovustatin treatment for 2 years results in a 41% decrease in lipid-rich necrotic core [12]. It should be mentioned, however, that despite the suggested decrease in lipid necrotic core, plaque volume remained unaffected in the latter MRI study. In another MRI study, ultrasmall supermagnetic iron oxide particles (USPIOs) were administered in patients suffering from greater than 40% stenosis of the carotid artery. USPIOs have a similar size to LDL particles and accumulate in the subendothelial region when it is permeable. The USPIOs can be visualized with MRI and are considered a surrogate measure for the presence of inflammatory cells because the USPIOs colocalize with intraplaque macrophages. High-dosage atorvastatin treatment resulted in a decrease in USPIO-defined inflammation while low-dosage treatment failed to show a similar effect [13].

Most human pharmaceutical intervention studies have demonstrated that, together with the anti-inflammatory parameters, LDL levels also decrease significantly when high dosages of statins are administered. Therefore, the study design does not always allow inferences to be made on LDL-independent pleiotropic effects of the statin treatment. In this article we will discuss a paper by Puato et al. that has recently been published in Stroke and reports the outcome of a study where the authors investigated the effect of two atorvastatin dosages and a nonstatin lipid-lowering drug [14].

Methods

A total of 60 patients scheduled for carotid endarterectomy were included, and 3 months prior to surgery patients were randomized to three groups. Patients either received a daily dose of 10-mg atorvastatin (AT-10 group), 80-mg atorvastatin (AT-80 group) or cholestyramine plus sitosterol (nonstatin control with LDL-lowering effect). A placebo group was not included for the aforementioned ethical reason. In all patients, inflammatory parameters such as CRP, IL-6 and IL-8 were measured in the blood at baseline as well as just before surgery. After surgery, plaques were examined, cut and stained for the presence of smooth muscle cells, macrophages and lymphocytes, and the amount of fat.

Results

At baseline, lipid parameters did not differ significantly among groups. However, a difference was observed at follow-up among groups for LDL and cholesterol levels. In the cholestyramine group, LDL levels were higher at follow-up compared with the atorvastatin-treated groups. Although data were not provided, the authors report that the circulating levels of proinflammatory cytokines did not change during follow-up. CRP levels showed a tendency towards a decrease in all three treated groups, but this was not significant, due to large variations. No differences in CRP levels were observed between the three groups at follow-up. Histological examinations revealed that smooth muscle cell content was highest in the AT-80 group. Moreover, the proportion of plaque area covered with macrophages was lowest in the AT-80 group. Although a tendency for a more stable plaque phenotype was still present in the AT-80 group, after adjustment for treatment-induced LDL changes the differences were no longer significant.

Discussion & conclusion

Based on the evidence provided, it was concluded that cellular plaque composition after short-term lipid-lowering therapy is modulated by the degree of LDL lowering, resulting in a decreased macrophage content, especially in the AT-80 group. It was hypothesized that the presence of fewer macrophages is likely to be associated with reduced risk of fibrous cap rupture. The authors explain why the design of the study is relevant to meet the objectives and discuss the limitations of other studies that either lack a nonstatin-treated control group or have retrospective design. They did not observe a lowering of proinflammatory cytokines or CRP in the circulation, which might be explained by the low number of patients included. This is an inherent drawback since recruitment of a larger high-risk and lipid-lowering naive population is limited due to the current standard of treatment. The results are in line with previous in vitro and animal studies that consistently show a pleiotropic effect of statin treatment.

Comment

Preclinical in vitro and animal experiments provide a body of evidence supporting the anti-inflammatory pleiotropic effects of statins. Human data on the plaque-stabilizing effects of statins are scarce and are hampered by low patient numbers and the lack of placebo-treated groups. In addition, when statin treatment was associated with an alteration of plaque phenotype, it could not be concluded whether this effect was directly the effect of a statin or an indirect effect of the LDL lowering. Puato et al. made an attempt to control for LDL levels by including a nonstatin-treated group that received cholestyramine. Unfortunately, but maybe expectedly, the patient group that did reveal a lower macrophage content in plaques during follow-up also demonstrated lower LDL levels in response to the treatment. Subsequently, the authors adjusted the outcome results for the lowering of LDL levels and report that the differences in macrophage percentages between groups no longer reached the significance threshold. Thus, part of the differences that were observed between treated groups could be explained by the lowering of LDL levels.

Based on this report, we cannot conclude that statins do not directly exert a pleiotropic anti-inflammatory effect on atherosclerotic plaques. A tendency towards lower macrophage content in the high-atorvasatatin group was still observed and the study is likely underpowered to answer the question whether statins exert anti-inflammatory effects independently of LDL lowering. CRP levels also did not significantly decrease during treatment. Large variations in CRP levels can be caused by nonspecific patient characteristics such as infections or tissue damage, and thereby distort the result. Therefore, CRP lowering can only be demonstrated in large series where the impact of outliers is limited, which supports the idea that the sample size may not have been sufficient to draw solid conclusions regarding the LDL-independent anti-inflammatory effect of statins in atherosclerotic plaques.

There are several factors that merit careful consideration when plaque stabilization is used as a surrogate measure of drug efficacy in human and animal models.

First, outcomes of pharmacological studies on atherosclerotic disease progression may differ substantially when animal models or humans are investigated. In animals, mostly disease initiation is investigated, while in humans plaques are only dissected when advanced disease is evident, which in most cases is complicated with a thrombotic event.

Second, patient numbers of positive studies are often small. It is well recognized that large effects are often reported by small studies while larger patient studies mostly reveal smaller differences in a measure of drug effectiveness. This is logically explained by the larger variation in smaller sample sizes, but it may lead to publication bias. Larger studies using plaque specimens have been reported with conflicting results, but these were retrospective [15].

Third, plaque characteristics may be influenced by many factors that are rarely taken into account in analyses. For instance, plaque inflammation may change rapidly following a cerebral thrombotic event [16]. The authors of the article under evaluation were right to take this factor into account by starting the therapy at a fixed time point following the cerebral event, and thereby minimizing the risk of confounding by timing.

Relevance & conclusion

Statins are part of the standard treatment in all patients suffering from atherosclerotic disease. Whether the beneficial effects are explained by reduced inflammation or LDL lowering will not alter this. The question of whether the outcome effect is explained by the anti-inflammatory actions of the drug is still relevant for future studies. For instance, the use of surrogate biomarkers to predict drug efficacy may depend on the biological role of the biomarker of interest that changes in response to a given drug.

A review has recently been published in Atherosclerosis that provides an overview of studies that examined the effects of statins on plaque morphology [17]. The authors performed a systematic review of available literature and concluded that further research is needed to determine whether the beneficial effect of statins on plaque phenotype is LDL associated. We agree with the conclusion of this article that there are inherent limitations to the safe extraction of solid conclusions from most studies due to data heterogeneity and publication bias.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.