Reconstituted HDL: A Therapy for Atherosclerosis and Beyond

Andrew J Murphy; Jaye Chin-Dusting; Dmitri Sviridov


Clin Lipidology. 2009;4(6):731-739. 

In This Article

Human Studies

It has previously been demonstrated that infusion of rHDL into healthy humans is well tolerated and significantly raises the levels of HDL and apoA-I.[31] It has also been shown that a single bolus infusion of rHDL over 4 h of 25 or 40 mg/kg of body weight raises circulating HDL levels to approximately 30 and 50 mg/dl, respectively, over baseline concentrations. These levels were significantly elevated for at least 25 h.[31] During the infusion, pre-β HDL peaked and conversion to an α-migrating HDL species occured within 2 h and baseline levels of pre-β migrating HDL were observed by 6–10 h, suggesting the rapid acquisition of cholesterol/lipid by these infused particles (i.e., inducing cholesterol efflux).[31]

An early risk factor for CVD is hypercholesterolemia, which is associated with a dysfunctional endothelium. Forearm blood-flow (FBF) measurements indicate that there is a diminished vasodilatory response to acetylcholine (ACh) in hypercholesterolimic patients that can be restored upon infusion of 80 mg/kg of rHDL.[32] Infusion of rHDL in Type 2 diabetics also reversed endothelial dysfunction. In both studies, the restoration in FBF by rHDL was attributed to an increase in NO as the NO inhibitor L-NMMA returned FBF to basal conditions. HDL has previously been shown to stimulate NO release in vitro by initiating a signaling cascade resulting in the phosphorylation of endothelial nitric oxide synthase.[33,34]

Potent antiatherosclerotic properties of rHDL were observed in animal models of atherosclerosis where results suggested that atheroma regression is achievable, or at the very least remodeling of the plaque to a more stable phenotype, paving the way for expansion of these studies to small human trials. Nissen and colleagues examined the antiatherosclerotic effects of a complex of rHDL featuring apoA-IMilano (ETC-216, Pfizer, UK) in patients with acute coronary syndromes.[15] The objective of the study was to determine the effect on plaque burden as measured by intravascular ultrasound (IVUS). Patients received five weekly infusions of placebo or ETC-216 at 15 mg/kg or 45 mg/kg. Baseline IVUS was performed within 2 weeks of an acute coronary event and repeated after the treatment period. The major finding of this study was that infusion of ETC-216 in the combined group reduced atheroma volume by 4.2% compared with baseline over the 5-week period.[15]

A follow-up clinical study was conducted in 2006 by Nicholls and colleagues to determine the relationship between atheroma regression and arterial wall remodeling based on animal studies, which provided good evidence that remodeling occurs as a result of rHDL infusion. The treatment groups remained the same as in the 2003 trial conducted by Nissen et al. and IVUS was again utilized to measure plaque burden. In addition, external elastic membrane (EEM) and lumen volumes were also compared. Moreover, there was a reduction in atheroma volume and a 4.6% reduction in EEM, whilst the lumen size was unaffected.[35] The investigators concluded that at sites where infusion of ETC-216 caused regression of coronary atherosclerosis, there was also reverse remodeling of the EEM without any change in lumen size.

Recently, a new version of rHDL has been developed (CSL-111, CSL Group, Australia). The ERASE trial was conducted to examine the effects of CSL-111 on coronary atherosclerosis, assessed by IVUS as in the rHDLMilano studies, where patients received four weekly injections of either placebo (saline), 40 mg/kg of CSL-111 or 80 mg/kg of CSL-111.[29] Unfortunately, the higher dosage group was discontinued owing to abnormal liver function tests. Whilst regression in atheroma volume between the 40 mg/kg CSL-111 group and the placebo group was not statistically different (−3.4 vs −1.6%), the reduction of 3.4% was significant in comparison with the baseline measurement in the 40 mg/kg cohort.[29] This study provided evidence for small, yet positive, effects of HDL infusion on atheroma regression, although longer term studies are required to assess the validity of such infusions as clinically useful. It must be recognized that the aforementioned studies employed IVUS (ETC-216 or CSL-111) as a surrogate measure of atherosclerosis and did not investigate clinical end points. It has yet to be determined if rHDL infusions in humans can cause meaningful changes in cardiovascular outcomes, and if so, what the exact indications for rHDL infusion are. It is not known whether rHDL can induce regression of the atherosclerotic plaque or halt its progression. Furthermore, no soluble markers of inflammation were examined in any of the IVUS studies; therefore, it is unclear how the potential changes in atheroma volume relate to the inflammatory status.

Data obtained from initial human clinical studies suggest that atheroma regression is evident and remodeling may be occurring. In addition, animal studies have demonstrated that acute increases in HDL have an effect on plaque composition. We endeavored to investigate if these findings translated into a more stable plaque phenotype in patients with peripheral vascular disease (PVD). Along with changes in plaque characteristics, we aimed to determine whether rHDL infusion had an effect on plasma markers of inflammation that are associated with atherosclerosis.[36] In a cohort of patients suffering with PVD, a single dose of 80 mg/kg of CSL-111 or placebo was administered and the patients underwent an atherectomy 7 days postinfusion. We observed a significant reduction in monocyte activation, which was in conjunction with a trend for a decrease in plasma markers of inflammation in patients that received the rHDL infusion. Changes observed in plaque composition included a significant reduction in lipid content, macrophage infiltration and VCAM-1 positive cells.[36] We have previously demonstrated that both HDL and apoA-I potently reduce monocyte activation, adhesion and migration in vitro, which could in part explain the observations of reduced macrophage infiltration in the excised plaques.[37] Thus, a single infusion of rHDL was shown to acutely alter plaque composition to what may be considered a more stable phenotype. However, these findings do require follow-up studies in larger cohorts and it would be interesting to determine if the changes in plaque composition in PVD patients are mirrored in patients suffering from central CVD, in particular, as the untreated plaques between these patients groups are likely to differ in their initial composition. It would also be important to determine how rapidly the rHDL is acting on the plaque and how long these effects are sustained, particularly as apoA-I levels have been reported to return to baseline levels shortly after infusion of rHDL.[38]

The efficiencies of ETC-216 (rHDL based on apoA-IMilano) and CSL-111 (rHDL based on wild-type apoA-I) were never compared head-to-head, and meta-analysis does not provide evidence that one may be more efficient than another. Studies in animal models[23,24] and in vitro[39] revealed that apoA-IMilano does not have superior cardioprotective functions over native apoA-I.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.