Plasma-Derived ApoA-1

SOME 800,000 American adults will experience a myocardial infarction (MI) this year.¹ Of the 85 percent who survive their heart attack, many are at high risk for a recurrent MI, stroke or other major adverse cardiovascular event, particularly over the first few months of their event.

Following acute-phase management of post-MI patients with drug and mechanical revascularization interventions, maintenance pharmacotherapy, cardiac rehabilitation and lifestyle modification prescriptions all help to moderate the risk of a recurrent cardiovascular event. But across all cases, there remains a roughly 10 percent risk of re-infarction, stroke and cardiovascular death over the first year following an MI.² Approximately two-thirds of these recurrent cardiovascular events occur in the first 90 days.

This high prevalence of early post-MI cardiovascular events clearly represents an important unmet medical need, and has attracted much interest and investment in potential protective treatments to reduce this recurrent disease burden. Most of these treatments target the key mediator of atherosclerotic disease itself: the low-density lipoprotein cholesterol (LDL) that builds up in the arterial walls, producing atherosclerosis that eventually can lead to MI and stroke.

Perhaps the most intriguing LDL-focused approach involves the development of products intended to mimic the physiologic activity of high-density lipoprotein cholesterol (HDL), the so-called “good cholesterol” that mediates the continuous removal, or “efflux,” of atherogenic LDL from vascular cells.

HDL and Reverse Cholesterol Transport

HDLs are actually a range of globular particles containing both proteins and lipids. The outer layer of HDL includes the more polar lipids, phospholipids and free cholesterol, while more hydrophobic lipids reside in the core of the particle. Apolipoprotein A-I (apoA-I), a 243-amino acid protein embedded in the outer layer, is the active mediator of cholesterol efflux, the first step in reverse cholesterol transport — a physiologic repair effort involving removal of atherogenic LDL from arterial cells and its elimination from the body.

The reverse cholesterol transport process occurs in three stages:

1) Cholesterol efflux mediated by apoAI, which removes excess cholesterol contained in lipid-laden cellular macrophages;

2) Lipoprotein remodeling, where HDL undergoes structural modifications with possible impact on its functional activity; and

3) Hepatic lipid uptake, where HDL releases cholesterol to the liver for excretion into bile and feces.

HDL particles can contain variable numbers of apoA-I protein molecules.³ Further, despite having similar or slightly higher apoA-I levels than healthy individuals, individuals with stable atherosclerotic disease exhibit significantly lower cholesterol efflux capacity, the functional measure of how effectively harmful cholesterol can be removed by HDL from vascular tissues. This suggests the physiologic status of these patients somehow may be altering their circulating HDL to render it less functional or even dysfunctional.⁴

Consistent with these findings, a study of more than 1,600 patients with ST-elevation myocardial infarction (STEMI) requiring percutaneous intervention found that those with the highest HDL-mediated cholesterol efflux capacity had markedly lower 30-day and long-term mortality rates following their MI event than those with the lowest efflux capacity, after adjusting for serum HDL level.5 HDL cholesterol efflux capacity is thus an inverse marker of incident cardiovascular disease risk.

It turns out the functionality of HDL in post-MI patients, entirely apart from the HDL serum level, may be key to successful development of HDL-based therapeutics to reduce major cardiovascular events following MI.

The Conceptual Basis for an HDL-Based Therapeutic

Data from the landmark Framingham Heart Study in the 1970s first established that, across a large population, the serum level of HDL cholesterol is inversely related to the risk of coronary heart disease (CHD). This independent protective effect of HDL may be even more potent than is LDL as a CHD risk factor: In the Framingham population, every 10 mg/dl increase in HDL was found to be associated with a roughly 50 percent reduction in CHD risk^6,7 (Figure 1). HDL is now incorporated into standard patient cardiovascular risk formulas — the ratio of total cholesterol to HDL, for example — to provide better predictive power for CHD.

This strong relationship between HDL level and cardiovascular event risk prompted two large clinical trials to learn whether oral niacin supplementation, which modestly elevates serum HDL by 15 percent to 20 percent in low-HDL patients with established atherosclerotic disease, could potentially reduce stroke or CHD-related hospitalizations, revascularization procedures or deaths.^8,9 Surprisingly, both studies failed to show a significant effect of niacin treatment on the incidence of major cardiovascular events. Several drugmakers later pursued testing of novel compounds with the capacity to more robustly boost HDL levels. Once again, large Phase III clinical trials failed to show long-term treatment with these more potent HDL-boosting drugs had a meaningful effect on the cardiovascular event rate compared to placebo.^10,11

Why would classical epidemiological studies show a strong association between higher HDL levels and lower cardiovascular event risk, while drugs used to increase circulating HDL levels fail to have a clearly demonstrable effect? The answer is thought to involve the distinction between simply raising serum HDL, particularly in people with existing atherosclerosis, and introducing more functional HDL particles that actually mediate reverse transport of cholesterol out of vascular tissues.¹²

Studies measuring cholesterol efflux capacity, again an independent predictor of cardiovascular event risk apart from the HDL level itself,^13,14 have revealed that blood sera from individuals with the same HDL levels can widely differ in their ability to promote the efflux of cholesterol bound up in macrophages.15 These and other findings have spurred interest in developing apoA-I — the key functional portion of HDL — as a potential therapeutic to boost cholesterol efflux capacity during the critical days and weeks following an MI, with the goal of reducing recurrent event rate.

Two recombinant apoA-I-based HDL mimetics have advanced from bench and preclinical testing to clinical trials, but once again with surprisingly disappointing results:

• ApoA-I Milano (MDCO-216) is a recombinant version of a high-functioning apoA-I variant isolated from residents of a village in Italy who have an unusually low prevalence of CHD. In a double-blind pilot study, 122 post-acute coronary syndrome patients placed on statin therapy were randomized to receive five weekly infusions of apoA-I Milano or placebo. At day 36, there was no significant difference in the mean change in atheroma volume between apoA-I Milano and placebo groups; this HDL mimetic failed to produce incremental plaque regression beyond the benefit produced by statin therapy alone.¹⁶

In late 2016, the sponsor (The Medicines Company) announced it discontinued further investment in the clinical development of apoA-I Milano/MDCO-216.

• CER-001 is an engineered complex of recombinant human apoA-I and phospholipids designed to mimic the structure and function of natural HDL. In a doubleblind trial, 272 patients with acute coronary syndrome and extensive coronary atheroma were randomized to receive 10 weekly infusions of CER-001 or placebo in addition to statins. Between baseline and day 78, mean atheroma volume decreased slightly in placebo group patients, but not at all in those in the CER-001 group; similar percentages in both groups demonstrated any measurable atheroma regression.¹⁷

The sponsor (ABIONYX Pharma) abandoned development of CER-001 for this clinical application in early 2017.

Australia-based CSL Behring has taken a different tack. In a research and development program now spanning well over a decade, the company has been investigating a novel formulation of apoA-I (CSL112) purified directly from donor human plasma. This naturally-derived apoA-I product has consistently produced strong, immediate and roughly comparable increases in cholesterol efflux capacity both in studies of healthy volunteers and in individuals with CHD.

CSL112: Human Plasma-Derived ApoA-I

Among the world’s leading manufacturers of human plasma-derived medicinal products, CSL Behring purifies intravenous and subcutaneous immune globulin, albumin and other therapeutics from millions of liters of donor plasma each year. In the early plasma fractionation process steps, HDL partitions to lipid-rich fraction IV. The native apoA-I protein is further purified from this fraction, lyophilized with specific stabilizers and reconstituted to form a preparation (CSL112) suitable for infusion.

In vitro studies of CSL112 have shown it undergoes remodeling steps that involve transient fusion with endogenous HDL and subsequent rapid fission to spontaneously yield two small-diameter and one large-diameter HDL molecular species, all containing CSL112. The smaller lipid-poor apoA-I HDL species demonstrates both highly functional cholesterol efflux capacity and significant anti-inflammatory effects.¹⁸

Like endogenous HDL, CSL112 acts as a scavenger, removing LDL from the coronary plaque and transporting it to the liver for eventual excretion. CSL112 was designed to optimize cholesterol efflux by the ATP-binding cassette transporter (ABCA1); this membrane “transporter” is induced by excess cellular cholesterol present in the atherosclerotic plaque (Figure 2).

Infusions of CSL112 into healthy volunteers have been shown to rapidly induce cholesterol efflux, the first step in reverse cholesterol transport. A single 6-gram dose of CSL112 induced a 2.5-fold average increase in total cholesterol efflux capacity compared to baseline, and was safe and well-tolerated in volunteers with both normal and moderately impaired renal function.¹⁹ Importantly, CSL112 performed similarly in persons with high and low endogenous HDL functionality. Further, pooled data from studies in 93 healthy subjects and 44 patients with stable atherosclerotic disease documented strong, quantitatively similar elevations in cholesterol efflux capacity, independent of disease status or baseline cholesterol efflux activity.²⁰

As worsening cardiovascular disease is associated with declining cholesterol efflux capacity, these findings suggested CSL112 can robustly boost cholesterol efflux capacity in patients with impaired endogenous HDL function. Encouragingly, several studies have documented the ability of CSL112 (as well as CSL111, a predecessor version) to reduce atherosclerotic plaque volume. In particular, a study randomizing 20 patients with femoral artery claudication, a single infusion of CSL111 resulted in acute changes in plaque characteristics and significantly reduced plaque lipid content and measures of inflammation compared to placebo.²¹

Assessing CSL112 in Post-MI Patients

Conducted in 2015 at treatment sites in 16 countries, CSL Behring’s Phase IIa AEGIS-I study randomized a total of 1,258 post-MI patients on a 1:1:1 basis to receive four weekly infusions of a low dose (2 grams apoA-I) or a high dose (6 grams apoA-I) of CSL112, or placebo. The higher dose of CSL112 was associated with well over a 100 percent increase in cholesterol efflux activity (Table), similar to that achieved in patients with stable coronary artery disease. There were no significant alterations in liver or kidney function or other safety concerns.

At 12-month follow-up, the rate of major adverse cardiovascular events — a composite of cardiovascular death, nonfatal MI, ischemic stroke and hospitalization for unstable angina — was similar between the low- and high-dose CSL112 treatment groups and the placebo group (respectively 6.4 percent, 5.7 percent and 5.5 percent). Nonfatal MI rates were the same — about 3 percent — in all three treatment groups. Two and four cardiovascular deaths were reported in the 2- gram and 6-gram CSL112 groups, and none in the placebo group; this one-year mortality rate was considered too small to be evaluable.

In early 2018, CSL Behring initiated its pivotal Phase III AEGIS-II study randomizing adult patients diagnosed with either STEMI or non-STEMI to receive four consecutive weekly infusions of 6 grams (170 mL) of CSL112 or a 4.4 percent albumin placebo solution. Study participants may be either medically managed or managed with percutaneous coronary intervention. Some 1,000 U.S. and international study sites are participating in this Phase III, double-blind, randomized, placebo-controlled, parallel-group study. With a 17,400-subject enrollment target, this is by far the largest and most ambitious clinical investigation of any plasma-derived therapeutic in history. AEGIS-II is scheduled to be completed in about two years.

An Opportunity to Make a Difference

CSL Behring and its clinical collaborators across the globe are hoping CSL112’s demonstrated ability to strongly boost cholesterol efflux capacity and reduce lipid content in atherosclerotic plaques will translate into a clinically meaningful reduction in a composite of fatal and nonfatal cardiovascular events over the critical 90-day period following MI.

For its pivotal AEGIS-II study, the company is focusing on a higher-risk subpopulation of MI patients with multivessel coronary artery disease and at least one of the following: age greater than 65 years, a history of prior MI, diabetes mellitus and/or peripheral artery disease. In this “enriched” AEGIS-II study population, CSL Behring anticipates a post-MI recurrent cardiovascular event rate of about 10 percent over the first 90 days in its placebo control arm.

An interim efficacy analysis will be completed next year. If analysis of the final results confirm CSL112 can meaningfully reduce that 90-day recurrent event rate with an acceptable safety profile, the product will immediately establish itself as a standard treatment for this high-risk segment of the 800,000 Americans who experience an MI each year.

The AEGIS-II trial is now past the halfway point. CSL Behring will spend roughly $800 million altogether to learn if its faith in CSL112 was justified. With thoughts of countless patients whose weeks and months following their heart attack are lived in fear, the medical community is hopeful too that CSL112 turns out to be a winning bet.