New lipid-lowering drugs: an update
Disclosures ASW served on the advisory board for Amsterdam Molecular Therapeutics from 2006 to 2008 and has received lecture honoraria from Aegerion Pharmaceuticals (2008). He has been an investigator on clinical trials for AstraZeneca, Merck Sharp & Dohme, and Novartis. AV and DM have received lecture honoraria, travel funds and commercial clinical trial support from AstraZeneca, Boehringer-Ingelheim, Merck Sharp & Dohme, and Takeda. TH is managing director of a commercial medical writing company but received no commission for this work.
Dr A. S. Wierzbicki,
Department of Chemical Pathology,
St. Thomas’ Hospital,
Lambeth Palace Road,
London SE1 7EH, UK
Tel.: 0171 928 9292ext2027
Fax: 0171 928 4226
Lipid lowering is established as a proven intervention to reduce atherosclerosis and its complications. Statins form the basis of care but are not able to treat all aspects of dyslipidaemia. Many novel therapeutic compounds are being developed. These include additional therapeutics for low-density lipoprotein cholesterol, for example, thyroid mimetics (thyroid receptor beta-agonists), antisense oligonucleotides or microsomal transfer protein inhibitors (MTPI); triglycerides, for example, novel peroxosimal proliferator activating receptors agonists, MTPIs, diacylglycerol acyl transferase-1 inhibitors and high-density lipoprotein cholesterol (HDL-C), for example, mimetic peptides; HDL delipidation strategies and cholesterol ester transfer protein inhibitors and modulators of inflammation, for example, phospholipase inhibitors. Gene therapy for specific rare disorders, for example, lipoprotein lipase deficiency using alipogene tiparvovec is also in clinical trials. Lipid-lowering drugs are likely to prove a fast-developing area for novel treatments as possible synergies exist between new and established compounds for the treatment of atherosclerosis.
Systematic literature search of clinical, scientific and financial sources for major therapeutic targets. Identification of compounds for each target in human clinical studies.
Message for the Clinic
Multiple novel therapeutics targeting LDL-C, HDL-C and inflammation are in development to modify lipid parameters and reduce atherosclerosis.
The progression of atherosclerosis, which is responsible for coronary heart disease (CHD), stroke, carotid and femoral artery stenosis (peripheral vascular disease), is associated with multiple cardiovascular disease (CVD) risk factors including hyperlipidaemia (1–3). A recent genetic study in man suggests that there are 95 genetic loci associated with lipid metabolism in man, all of which are potential drug targets (4). In epidemiological studies a 1% reduction in low-density lipoprotein cholesterol (LDL-C) is associated with a 1% reduction in cardiovascular events and a 1% increase in high-density lipoprotein cholesterol (HDL-C) is associated with a 3% reduction in events (5). The data on triglycerides, triglyceride-rich remnants, small dense lipoprotein subfractions and lipoprotein (a) are less clear but all are associated with increased risk (6–8). Statins are acknowledged first-line drugs for most lipid disorders but do not address all CVD disease risk. At least part of this ‘residual risk’ may be attributable to dyslipidaemia (9). Common use of statins has also exposed a group where their use is limited by side effects. Discontinuation of therapy is seen in about 5% of patients receiving statins or fibrates and 30% in the case of bile acid sequestrants. The importance of hyperlipidaemia as a problem means it is an active area for drug development. Many of the targets identified almost a decade ago still remain (10) but new themes have emerged with the discovery of preprotein subtilisin kexin 9 (PCSK-9) as a control system for the LDL receptor and with the development of antisense oligonucleotide (ASO) technology.
This review was based on PubMed searches for each major drug class listed for trials of compounds in man for hyperlipidaemia (cholesterol, triglycerides, LDL-C, HDL-C) allied with searches of ClinicalTrials.gov, medical abstracts services (MedScape), financial analytical reports on the lipid-lowering drugs (Reuters, Forbes, DataMonitor), and individual pharmaceutical company websites and press releases up to 14 November 2011.
New variants on old themes
Little has changed since the previous review of this topic. Statins are effective and safe drugs that can produce a maximum LDL reduction of about 60% with parallel reductions in triglyceride and a modest rise in HDL through independent mechanisms (11). Their main disadvantage is that dose titration by doubling only results in a 5–7% extra reduction in LDL but at the expense of increasing side effects especially myalgia-myositis. The clinical evidence for the use of statins in both secondary and primary prevention of CVD is overwhelming. Atorvastatin and rosuvastatin are well established and seem to differ little even at the highest doses as the results of the Study of Coronary Atheroma by Intravascular Ultrasound: Effect of Rosuvastatin vs. Atorvastatin (SATURN) study using intravascular ultrasound (IVUS) in 1385 patients show (12). Pitivastatin is licensed in a few countries (13). Statins in development include NCX-6560 (NicOx; Sophia-Antipolis, France), a statin with added vasodilator (nitric oxide donor) properties which reduces LDL-C by 57%(14) and PPD-10558 (Furiex; Morrisville, NC) a statin with enhanced liver extraction and reduced plasma exposure now being trialled in statin-intolerant patients (15).
Fibrates are an established drug class that act as nuclear agonists at the peroxisomal proliferator activating receptor alpha (PPAR-α) element that regulates expression of apolipoprotein A1 and A2 (principal components of HDL), lipoprotein lipase, apolipoprotein C-3 (associated with triglyceride-rich remnant particles), acute phase proteins including fibrinogen and C-reactive protein (CRP) and clotting factors (plasminogen activator inhibitor 1) amongst many other genes. Their chief actions are to lower triglycerides and to slightly raise HDL-C. There is evidence that fibrates reduce CVD events by a small amount (10–13%) in monotherapy (16) and in patients with low HDL and elevated triglycerides may add to baseline statins although this still requires formal large scale clinical trial confirmation (17,18). Fibrates show differential specificity for tissue PPARs with a predominant PPAR-α action. They also seem to have microvascular as well as macrovascular effects although whether or not this is related to their lipid-lowering efficacy is debatable (19). Newer, more specific PPAR-α agonists (e.g. FF-200) were developed but failed because of increased muscle toxicity and the trend seems to be to investigate multi-PPAR active moieties (see below).
Bile acid sequestrants
Bile acid sequestrants reduce cholesterol absorption by sequestering bile acids and preventing opsonisation of lipid-rich particles in the gut by bile acids and cause secondary up-regulation of hepatic LDL-receptors. They raise triglycerides through effects on the liver-X-receptor (LXR), have minimal effects on HDL and also reduce glucose through a farnesoid-X receptor action. Their main limitation is a high side effect rate caused by the presence of colonic bile acids resulting in bloating, nausea and diarrhoea. Colesevelam is the latest version and has licence for improving glycaemia as well as reducing lipids (20).
Niacin is the oldest of the lipid-lowering drugs. Its mechanisms of action is not clear but involves reducing the synthesis of very low-density lipoprotein (VLDL) in the liver through effects on diacylglycerol acyl transferase-2 (DGAT-2) as well as affecting peripheral lipolysis through the FM-70 receptor (21). It raises HDL-C by 20–25% through increases in aPoA1 production and a reduction in clearance through the ATP-synthase β-chain holoparticle receptor. It is one of the few agents that can reduce Lp(a) concentrations. Studies including the Coronary Drug Project show a 20% reduction in CVD events with niacin monotherapy (22) but recently in the underpowered AIM-HIGH study it failed to add to aggressively used baseline LDL-lowering therapies (23). Its use has been limited by the flushing caused by its activation of prostaglandin E synthesis and hepatotoxicity but the addition of laropiprant (a prostaglandin D2 type 1 receptor antagonist) reduces the incidence of flushing by 70–80%. The effect of the combination of niacin and laropiprant on CVD events is being investigated in the Heart Protection-3/Treatment of HDL to Reduce the Incidence of Vascular Events (HPS-3/THRIVE) study (24). A variety of non-flush niacin derivatives have been synthesised but only ARI-3037MO (Arisaph Pharmaceuticals; Boston, MA) has reached clinical trials in man (25).
New classes of drugs: therapies to reduce LDL-cholesterol – drugs acting on lipoprotein synthesis
Squalene synthase inhibitors
Although early and the final stages of cholesterol synthesis occur in peroxisomes, one of the limiting stages of cholesterol synthesis occurs in the endoplasmic reticulum and is catalysed by squalene synthase. Inhibition of squalene synthase leads to a reduction in cholesterol synthesis without affecting synthesis of compounds derived from geranyl pyrophosphate including dolichols and ubiquinone (co-enzyme Q10) or affecting the control of protein function through farnesylation. It thus offered potential to reduce myalgia (potentially linked to CoQ10 levels in muscle) but possibly not to reduce inflammation mediated via prenylation and rho kinase (26). Lapaquistat (TAK-475; Takeda; Osaka, Japan) unfortunately caused hepatic dysfunction and had only small effects (∼15%) in reducing LDL-C and so was discontinued (27).
Microsomal transfer protein (MTP) inhibitors
Statins increase hepatic LDL receptor expression, and hence enhance plasma clearance of LDL. There is also some data to suggest they have an effect on reducing the synthesis of VLDL through reducing the size of the cholesterol sub-compartment in endoplasmic reticulum necessary for stabilising apolipoprotein B100. The direct link between this putative pool and the ER is provided by MTP (28). Specific inhibitors of MTP have been shown to deliver LDL reductions of 70–80% with a reduction in triglycerides of 30–40% (28). Unfortunately, inhibition of MTP results in an increase in the size of hepatic cholesterol pools and the development of fatty liver (non-alcoholic steatitic hepatitis) and elevated transaminases through an unopposed action of fatty acid synthases. Of this class only lomitapide (AEGR-733; Aegerion Pharmaceuticals, Cambridge, MA) survives. This is a low dose version of BMS-201038 which proved effective in reducing lipids in patients with homozygous familial hypercholesterolaemia (FH) but at the expense of causing liver transaminitis in 33% (29). Further trials of lomitapide in patients maintained on very low fat diet and/or on the background of ezetimibe which itself may reduce liver fat proved its efficacy in monotherapy and in combination with ezetimibe in reducing lipids in a mild mixed hyperlipidaemic population (30). Although likely to obtain an orphan drug licence in homozygous FH the main use of lomitapide may be in gross hypertriglyceridaemia. Previous studies with microsomal transfer protein inhibitors (MTPIs) have shown that they are effective in reducing both VLDL and chylomicron production and that specifically gut-targeted MTPIs (e.g. SLX-4090; Nano Terra, Brighton, MA) can reduce postprandial triglycerides (28). Thus, the main potential for this drug may be in the treatment of type V or type I hyperlipidaemia.
Acyl-cholesterol acyl transferase (ACAT) inhibitors
One of the critical steps in atherosclerosis and in cholesterol uptake is the esterification of cholesterol in macrophages that promotes foam cell formation and is required for lipoprotein synthesis (31). Avasimibe and Pactimibe which inhibited both ACAT-1 and 2 reached clinical trials but failed to reduce atherosclerosis progression in IVUS studies (32,33) or carotid intima-media thickness (IMT) in patients with FH (34) and development of these drugs stopped. The ACAT-1 selective K-604 (Kowa Pharmaceuticals; Nagoya, Japan) remains in development (35).
The Coronary Drug Project included dextro (d)-thyroxine as one of its treatment options. Although dextro (d)-thyroxine reduced LDL-C in the CDP it increased rates of arrhythmias and that arm of the trial was discontinued (36,37). Further researches clarified the role of thyroid α- and β-receptors and have suggested that a β-agonist would reduce lipids without any adverse general metabolic or cardiac consequences. Eprotirome (KB-2115; Karo Bio AB; Stockholm, Sweden), sobetirome (GC-1; QRX-431; QuatRx Inc, Ann Arbor, MI) and T-0681 are thyroid β-receptor agonists that in the case of eprotirome reduce LDL-C by 7–32% in man without causing significant hepatic or muscle dysfunction (38). It partially seems to act by modulating LDL receptor expression in the liver (39) and is efficacious in heterozygous FH but on this basis may not work in homozygotes. However, like niacin but unlike other current lipid-lowering drugs, it also reduces lipoprotein (a) by 25% indicating that is has effects on lipoprotein synthesis as well (40).
Antisense oligonucleotide therapies (LDL-C)
It is possible to specifically reduce or silence gene expression by infusing a short complementary ASO sequence to messenger RNA (41). Silencing through the formation of double stranded RNA is well known and forms a mechanism of control of gene expression. However, these interfering RNAs are unstable; but, it is possible to make stable oligonucleotide derivatives that are protected against hydrolysis either by addition of blocking groups and/or by fixing the strand conformation (42). These changes and addition of specific domains, for example, asialoglycoprotein binding can modify the clearance of ASOs to make them liver-specific as opposed to requiring the scavenger receptors (43). Now numerous individual methods exist to make stable ASOs. The potential for treatment of lipid disorders was noted by considering analogies with genetic hypo-betalipoproteinaemia where reduced levels of apoB are associated with highly reduced LDL-C but normal lifespan (28). Mipomersen (ISIS 301012; Isis, Carlsbad, CA/Genzyme, Cambridge, MA) is the first human antisense therapy. In humans it reduces apoB and LDL-C in a dose proportional manner by 20–65% (44,45), is effective in homozygous and heterozygous FH in reducing LDL-C by 25% when added to standard baseline therapies and has a reasonable safety profile (46). Mipomersen also reduces levels of Lp(a) by 25% likely through reducing its synthesis (40). Its main side effects are florid injection site reactions and hepatic steatosis (47,48). Hepatic steatosis is a common finding in patients with hypobetalipoproteinaemia and does not seem to be associated with progression to non-alcoholic steatohepatitis and liver cirrhosis (49). Mipomersen is under consideration for licensing and is likely to find its first use in patients with severe FH undergoing apheresis where it may enable them to stop this highly invasive treatment.
The ASOs have the potential to be applied to any drug target and many animal studies exist in the field of hyperlipidaemia. Beyond apoB only ISIS-ApoC-III (Isis) is in clinical development for treatment of hypertriglyceridaemia given the potential for this treatment to improve triglyceride clearance (50,51).
Preprotein convertase subtilisin kexin-9 inhibitors
Preprotein convertase subtilisin kexin-9 is a newly discovered protein involved in intracellular and extra-cellular regulation of LDL receptor expression (52). In general, autosomal dominant activating mutations in PCSK-9 cause FH whereas inactivating mutations have been shown to be associated with 0.3–0.5 mmol/l reductions in plasma LDL-C and 70–80% lifetime reduced risks of CHD events (53) but exceptions exist (54). PCSK-9 activity is related to fasting, postprandial lipid metabolism and hormonal control of lipids by oestrogens, androgens and growth hormone (55). PCSK-9 effects account for some of the LDL-lowering effects of fibrates (56) and berberine (57). One patient with homozygous PCSK-9 deficiency has been described and is hypocholesterolaemic but supposedly fit and well  but mice with PCSK-9 deficiency are hyperglycaemic, hypoinsulinaemic and had abnormal pancreatic islets (58,59). Antibodies to PCSK-9 increase LDL-R expression and reduce plasma LDL-C by 20% (60). Numerous immunisation therapies to PCSK-9 are in development and include the human monoclonals REGN-727/SAR-236553 (Regeneron, Tarrytown, NY/SanofiAventis, Paris, France) (61), and AMG-145 (Amgen; Thousand Oaks, CA) (62) as well as with humanised antibodies, for example, NVP-LGT-209 (Novartis, Basel, Switzerland) (63). Recently REGN-727 was shown to reduce LDL-C by 36–58% in monotherapy or added to statin in a dose ranging safety trial in patients with FH (64) and similar results were seen in a dose ranging study of AMG-145 in healthy volunteers (65). Other potential approaches include small molecule inhibitors and ASOs (e.g. Alnylam Cambridge Massachusetts, ALN-PCS, Idera, Cambridge Massachusetts, ISIS BMS-PSCK-9Rx or Santaris Hørsholm, Denmark SPC5001).
Therapies to reduce LDL-cholesterol: drugs acting on intestinal lipid absorption
About 25% of cholesterol is derived from intestinal uptake. Treatment with sitostanol/sitosterol margarines that act as competitive inhibitors of cholesterol uptake show an 8–14% reduction in LDL-C if doses of 20 g/day are consumed (66). The results of the Program on Surgical Control of Hyperlipidemia (POSCH) show that ileal bypass surgery can produce a 35% reduction in LDL which is associated with a 24% reduction in cardiovascular events after 5 years indicating the possible maximum degree of achievable LDL reduction with inhibition of cholesterol uptake (67).
Cholesterol absorption inhibitors
Ezetimibe is a cholesterol uptake inhibitor which inhibits duodenal Niemann-Pick C1-like protein 1 (NPC1L1). Ezetimibe is well tolerated but like all lipid-lowering agents to date, causes nausea or bloating probably through transient bile acid/LXR metabolism dysfunction. It has an orphan drug licence for treatment of sitosterolaemia. In general use it has proved controversial with no evidence for benefit in initial surrogate marker studies (Ezetimibe and Simvastatin in Hypercholesterolemia Enhances Atherosclerosis Regression, ENHANCE) as monotherapy or in initial underpowered CVD end-point studies (Simvastatin and Ezetimibe in Aortic Stenosis, SEAS) in combination with a statin. Lately combination statin-ezetimibe therapy reduced CVD events by 27% in a renal failure population in the Study of Heart and Renal Protection (68). Add-on monotherapy data on CVD outcomes is still awaited from the IMProved Reduction of Outcomes: Vytorin Efficacy International Trial (IMPROVE-IT) study in acute coronary syndromes. Some small studies have suggested that ezetimibe may improve hepatic steatosis.
Bile acid sequestrants and ileal bile acid transport inhibitors (IBATIs)
Bile acid sequestrants, which inhibit the enterohepatic circulation of bile acids, are well established lipid-lowering drugs with additional benefits in reducing glucose. No novel agents have proceeded to clinical studies in man in a decade and there has been no progress with IBATIs.
Therapies for triglycerides
Omega-3 fatty acids
Both docosahexaenoic acid (DHA) and eicosapenatenoic acid (EPA) have been shown to reduce triglycerides in a dose dependent manner and have been used in a variety of preparations to reduce triglycerides (69). Both the GISSI-Prevenzione with DHA–EPA (70) and the Japan EPA Lipid Intervention Study (JELIS) (71) have shown their lipid-independent benefits in reducing cardiovascular events. Most recent research has focused on extending the triglyceride efficacy range of an EPA preparation (AMR-101; Amarin Pharmaceuticals, Dublin, Eire). In the Multicentre plAcebo controlled Randomised double blINd 12-week study with open label Extension (MARINE) studies further extensions of the dose-response relationship have been shown in patients with hypertriglyceridaemia with 33–45% reduction in triglycerides in patients with TG > 7.8 mmol/l (72). As a specific receptor for omega-3 fatty acids has been discovered (73) mediating their lipid, insulin sensitising and anti-inflammatory effects, the potential exists to develop specific new drugs for this target.
Diacylglycerol acyl transferases (DGATs) are involved in triglyceride synthesis in adipose tissue, the gut and in the liver (74). DGAT-2 may be one of the mechanisms by which niacin reduces hepatic triglyceride and hence VLDL production and is expressed in liver and adipose tissue (75). In contrast DGAT-1 is expressed in the intestine, liver and adipose tissue and data from DGAT-1 deficient mice and with inhibitors shows that it is a mechanism that mediates reduction in triglycerides, hepatic steatosis, obesity and improvement in insulin resistance. The DGAT-1 inhibitor LCQ-908 (Novartis) is in phase 2 trials in man and has been compared with sitagliptin on a background of metformin therapy (76) and is due to start trial in hypertriglyceridaemic patients.
Therapies to raise HDL
The HDL-C is consistently associated with a reduction in CVD risk and has multiple potential protective roles against atherosclerosis (77). A large literature exists showing the epidemiological benefits of a higher HDL-C possibly to an upper limit of 1.5 mmol/l (6,78). Cases of hyperalphalipoproteinaemia with higher levels (> 2 mmol/l) of HDL-C exist but have not been intensively studied with regards to CVD outcomes. Data from intervention trials are scarcer and mostly rely on drugs such as statin that have only small effects or on the few older trials of fibrates and niacin (79). A small benefit would be consistent with the results but fibrates and niacin also have other actions on lipid profiles that could confound these analyses.
Cholesterol ester transfer protein (CETP) inhibitors
Cholesterol esters are transferred in exchange for triglycerides in plasma by CETP both within groups of particles and between the apoB and apoA1-containing sub-sets (80). CETP deficiency occurs in 1 in 50,000 patients and is associated with HDL-C levels > 3 mmol/l mostly in the form of alpha-particle density proteins rich in apoE but deficient in apoA1. Patients also show reduced levels of LDL-C with the presence of an additional novel beta-lipoprotein band (81). In Okinawa, where it is common, patients with CETP deficiency do not show evidence of decreased atherosclerosis but studies where CETP activity has been measured suggest a small reduction in CVD events with decreased levels.
The first CETP inhibitor developed-torcetrapib (Pfizer, Groton, CT) was undoubtedly efficacious on lipids. It raised HDL-C by up to 130% and reduced LDL-C by 25% (82). However, in lipid turnover and tracer studies it showed no benefit on net cholesterol efflux to faecal sterols (83). In initial surrogate outcome studies torcetrapib showed little benefit on carotid IMT (Rating Atherosclerotic Disease change by Imaging with A New CETP Inhibitor, RADIANCE 1 & 2) or IVUS-quantified atheroma volume but subsequent analyses suggested that effects paralleled changes in LDL-C in the carotid IMT studies (84) and also that torcetrapib reduced HbA1c by 0.23% (85). In the Investigation of Lipid Level Management to Understand its Impact in Atherosclerotic Events (ILLUMINATE) CVD end-point trial torcetrapib raised CVD events by 30% leading to premature termination of its development programme (82). Its toxicity profile has been the subject of considerable investigation. Torcetrapib raised blood pressure by 5/2 mmHg partially through increased aldosterone, corticosterone and endothelin-1 production now proven off-target drug effects (86).
Subsequent compounds have undergone large scale safety testing as a result. Dalcetrapib (Hoffman La Roche, Basel, Switzerland) raises HDL-C by 25–33% with little effect on LDL-C (87). It does not cause hypertension or affect hormone profiles. However, both in preliminary trials on endothelial function (88) and recently in the larger 476 patient Dal-VESSEL trial no benefit was seen (89). Similarly in a combined positron emission tomography and magnetic resonance imaging-imaging Dal-PLAQUE trial in carotid atherosclerosis it failed to show significant benefits on plaque inflammation content, or plaque composition (90). The Dal-OUTCOMES end-point trial is underway in 15,000 patients with acute coronary syndromes (91) and it seems no adverse safety signals have been detected after a median follow-up of 2 years. Anacetrapib raises HDL by 90–130% and reduces LDL-C by 30–40% (91). Like dalcetrapib it had no effect on blood pressure or other safety parameters in the 1623 patient DEFINE trial over 2 years. The Heart Protection Study-3 (HPS-3) Randomized EValuation of the Effects of Anacetrapib Through Lipid-modification (REVEAL) outcome trial of anacetrapib added to baseline LDL-C therapy is underway in 20,000 patients with established CVD (93).
Other small molecule CETP inhibitors in development include evacetrapib (Eli-Lilly; Indianapolis, IN) (94), and DRL-17822 (Dr Reddy Laboratories, Hyderabad, India) (95). Evacetrapib has similar efficacy to anacetrapib increasing HDL-C by 54–130% and reducing LDL-C by 14–36% in monotherapy or added to statins. Vaccines against CETP (CETPI) have also been developed and were non-allergenic and reduced CETP but without a consistent effect on HDL-C levels in initial single injection trials in man. They have not proceeded in development.
Combined peroxisomal proliferator activating receptor alpha-gamma agonists
Atherosclerosis in type 2 diabetes is associated with low HDL-C, high triglycerides and hyperglycaemia. Similar but lesser changes occur in patients with the metabolic/insulin resistance syndrome. PPAR-α agonists (fibrates) can reduce triglycerides by 70%, raise HDL by 20% and reduce LDL by 10–25% but have small effects on HbA1c (18). PPAR-γ agonists (thiazolidinediones; glitazones) deliver a 0.5–1% reduction in HbA1c, 5–15% in triglycerides and raise HDL by 0–4% (96). PPAR-γ agonists have proved controversial as it was discovered that although they were effective in reducing HbA1c they caused aldosterone-driven fluid retention, excess fractures and in the case of rosiglitazone excess myocardial infarction (97). Pioglitazone, however, reduced CVD events by 12% in the PROspective PioglitAzone Clinical Trial In MacroVascular Events (ProACTIVE) trial (98). As PPARs share structural homology it has proved possible to synthesise PPAR α-γ co-agonists. Muraglitazar, tesaglitazar and ragaglitazar reached clinical trials but development was abandoned after excess bladder cancers were detected in animal models with ragaglitazar, MK-0767 (Merck), and naveglitazar and excess CVD events were noted with muraglitazar (99,100). The cancer effect seems to be a PPAR-γ driven effect as recent pharmacovigilance surveys have suggested a similar problem with pioglitazone in man (101). Only aleglitazar (R1439; Hoffman LaRoche, Basel, Switzerland) remains. Aleglitazar reduced HbA1c by 0.36–1.35%, triglycerides by 12–38%, and reduced LDL-C by 3–22% and raised HDL-C by 12–27% in 332 patients with type 2 diabetes in the SYNCHRONY dose-ranging (50–300 μg) without any increased heart failure weight gain or fluid retention at lower doses (102). The Aleglitazar Cardiac Outcomes (ALECARDIO) study in 6000 patients with type 2 diabetes and acute coronary syndromes is assessing the CVD effects of aleglitazar (103). It remains unclear given the availability of other agents for treatment of diabetes and the side effect profiles of PPAR-γ agonists whether or not any PPAR α-γ agonist will be successful. A PPAR-γ/δ agent DB-959 (Dara Biosciences; Raleigh, NC) is also in development and has reached phase I trials.
Although both PPAR-α and PPAR-γ agonists have beneficial effects on surrogate markers of diabetes and lipids as they seem to be limited in their clinical outcomes efficacy, attention has turned to the ubiquitously expressed PPAR-δ receptor. In contrast to PPAR-α (liver) and PPAR-γ (adipose) tissue effects PPAR-β/δ may be responsible for muscle lipid and glucose handling. In mouse studies PPAR-β/δ agonist improve lipid and glycaemic profiles. Recently lipoprotein turnover studies in man have confirmed that the PPAR-δ agonist GW501516 had analogous effects to a fibrate on VLDL turnover in increasing triglyceride-rich lipoprotein clearance through the apoB-100 LDL receptor and by reducing apoC-3 levels (104). In early studies in patients with mixed hyperlipidaemia or prediabetes (metabolic syndrome) the PPAR-α-δ co-agonist GFT-505 (Genfit, Loos, France) reduced triglycerides by 17%, raised HDL-C by 8%, and improved insulin resistance and gamma-glutamyl transferase levels (105).
Other HDL-raising drugs
Other methods of raising HDL-C exist besides known PPAR agonists. A quinazoline RVX-208 (Resverlogix; Calgary, Alberta, Canada) increases apoA1, pre-beta HDL and plasma HDL-C by 5–11% in monkeys and man in monotherapy and when added to statins (106,107). Further development is underway.
HDL-derived proteins, peptides and infusions
Point mutations in apoA1, the principal protein component of HDL are associated with a wide variety of clinical phenotypes including amyloidosis, neuropathy and both increased and decreased rates of atherosclerosis. ApoA1 Paris and Milano (R151C; R173C, respectively) are associated with reduced plasma HDL levels, increased reverse cholesterol transport and protect against atherosclerosis in man and animal models. Pure apoA-1 protein shows no benefit and is degraded in the kidneys. However, pre-HDL discs containing apoA1-Milano and phospholipid are functional, non-antigenic, not instantly cleared and reduce coronary atherosclerosis in preliminary IVUS studies in man (108). However, the dose response relationship was not linear in the IVUS trial although it did correlate with HDL rise (109). A subsequent infusion trial of reconstituted HDL (rHDL) CSL-111 (CSL Ltd, Parkville, Victoria, Australia) failed to show significant benefits but may have suffered from under-dosing and a too short trial period, given the lesser activity of rHDL as compared with apoA1MilanoHDL but studies are under way with CSL-112(110). Manufacturing problems limited the development of ETC-216 but plant-synthesised des1,2 apoA-1Milano retains activity and may allow further development. Numerous other HDL mimetic peptides have been developed including APP-018 (d4F) which was effective in mice (111) but not bioavailable in man. The ApoA1 mimetic peptide ATI-5261 (Advanced Therapeutics Inc, Clinton, MI) increases faecal cholesterol excretion in mice (112) and may reach clinical trials in man. Another ApoA1-derived phospholipid-loaded HDL mimetic CER-001 (Cerenis Therapeutics SA, Labege, France) given as weekly infusions is under evaluation in acute coronary syndromes using IVUS atheroma volume as an end-point in the Can HDL Infusions Significantly Quicken Atherosclerosis Regression (CHI-square) trial (113).
Another approach that can be taken is to modify the apheresis process of lipoprotein removal. Apheresis and plasmapheresis both reduce progression of atherosclerosis in patients (112) with homozygous FH. Instead of removing apoB-containing lipoproteins (e.g. LDL) it is possible to selectively remove apoA-1 (HDL) particles, delipidate them and re-infuse the cholesterol-depleted functional pre-beta HDL and use it to reduce atheroma in monkeys (114). Delipidated HDL has been shown to reduce IVUS-measured carotid atheroma volume after only a few infusions in patients with acute coronary syndromes (115) and machine-based delipidation is now a large scale trial in acute coronary syndromes as an acute intervention.
Inhibitors of inflammation-associated markers
Phospholipases are involved in the pathogenesis of atherosclerosis and phospholipase A2 is found as both secretory forms (sPLA2) and associated with lipoproteins (LpPLA2). Increased levels of LpPLA2 are associated with increased CVD risk and mechanistically linked to the production of lyso-phosphatidyl-choline a key signalling molecule in atherosclerosis (116). Inhibitors to both LpPLA2 (darapladib; GlaxoSmithKline, Brentford, London, UK) and sPLA2 (varespladib; Anthera Pharmaceuticals; Hayward, CA) are in development. Darapladib showed no effect in an IVUS study in man on general markers of inflammation (e.g. C-reactive protein), arterial stiffness or atheroma volume in the IBIS-2 study (117). Virtual histology suggested a reduction in necrotic core progression in darapladib-treated patients but no increase in plaque fibrosis was seen. The STABILITY (118) and SOLID/TIMI-52 trials (119) in 13,000 patients with CHD are investigating the clinical benefits of darapladib therapy.
The Fewer Recurrent Acute coronary events with Near-term Coronary Inflammation Suppression-ACS (FRANCIS-ACS) study investigated the safety and efficacy of varespladib (A-002; Anthera Pharmaceuticals) in 625 patients with acute coronary syndromes and showed an 8% reduction in LDL-C and 70–80% reductions in CRP and sPLA2 levels (120). The PhosphoLipase And Serological Markers of Atherosclerosis-2 (PLASMA-2) study showed an improvement in LDL subfractions (121) but no benefit was seen on myonecrosis with pretreatment prior to percutaneous coronary intervention in the SPIDER-PCI trial (122). The Vascular Inflammation Suppression to Treat Acute Coronary Syndrome-16 Weeks (VISTA-16) trial is investigating the role of varespladib added to statins in 6500 patients with acute coronary syndromes (123).
Direct inhibitors of atherosclerotic plaque signaling
Expression of inter-cellular adhesion molecules in response to the presence of oxidised cholesterol is one of the earliest steps in endothelial activation, which is the initial process in atherosclerosis. For many years probucol was used to treat atherosclerosis without any clear idea of its mechanism of action although it reduced LDL-C by 17% and HDL-C by 23%. It did not change mean femoral lumen diameter but it did reduce restenosis events post-angioplasty in Probucol Quantitative Regression Sweden trial (124). A derivative succinobucol (AGI-1067; Atherogenics; Alphareta, GA) failed to reduce CVD events including interventions in 6144 patients with acute coronary syndromes in the ARISE trial but did reduce the secondary end-point of CVD events by 19% with a reduction in HbA1c but with an excess of atrial fibrillation and heart failure (125). No further development of AGI-1067 has occurred.
Gene therapy was originally attempted in homozygous FH but was unsuccessful (126,127). Since then few attempts have been made to restart these studies in lipid disorders despite the progress made in improving viral vectors. However, recently an adeno-associated virus-based treatment for lipoprotein lipase deficiency using intra-muscular injection of an enhanced activity variant of LPL (S447X)-alipogene tiparvovec was attempted (AMT-001; (Amsterdam Medical Therapeutics, Amsterdam, the Netherlands) (128). In trials in homozygous LPL-deficient patients in Quebec and Holland this treatment transiently reduces triglycerides but does reduce the incidence of pancreatitis by 75% (129).Using tracer studies its effects seem to be mediated by a reduction in de novo chylomicron synthesis (130,131). Recently the European Medicines Evaluation Agency gave an unfavourable opinion on the risk-benefit of alipogene tiparvovec and development was discontinued.
This review has summarised some of the compounds in development for treatment of lipid-related risk factors in CVD. The success of statins will be a major factor in the uptake of alternatives. Statins are safe, available as generics and likely to be available over-the-counter either as monotherapy or as a part of a poly-therapeutic package [e.g. the poly-pill (132) or PolyCap (133)]. Thus, newer agents will have to show significant advantage in tolerability, safety and efficacy over existing agents to be used in monotherapy and even they are likely to be used in rare ‘niche’ indications. Combination therapy is well established for bile acid sequestrants, nicotinic acid or fibrates with statins and is likely to increase as lipid management is tailored to individual profiles especially as all trials will be conducted on baseline statin. Safety and efficacy studies will be necessary to prove the benefits of combinations involving new compounds as recent trials of established combinations using fibrates and niacin have been disappointing in terms of demonstrating increased benefits. In the meantime targets are likely to be re-set. Those for LDL-C may be lowered further for high-risk cases (e.g. < 60 mg/dl; 1.5 mmol/l) and targets will be promulgated for HDL-C (e.g. > 40 mg/dl, 1 mmol/L (5) and maybe 55 mg/dl; 1.4 mmol/l) and triglycerides (e.g. <150 mg/dL; <1.70 mmol/L or <200 mg/dl: <2.3 mmol/L) may be added on the basis of ongoing trials (6–8).
Much development still focuses on classical inhibition of enzymes and receptors although novel approaches use gene technology to accomplish this. Although some opportunities remain, for example, altering the apoE/C-3 balance to increase clearance of triglyceride-rich lipoproteins flux, it could be argued that the principal requirement in atherosclerosis is for agonists for macrophage cholesterol efflux, reverse cholesterol transport (i.e. increased turnover and not just plasma HDL-C) and enhancement of LDL-C clearance (e.g. paradoxically by inhibiting PCSK-9). Some of these strategies may not change plasma lipid concentrations and it will be necessary to investigate their utility using active plaque imaging [e.g. FluoroDeoxyGlucose positron emission tomography (90) or ultra-small particle iron oxide (USPIO) NMR], virtual histology and other imaging techniques prior to proving efficacy in end-point studies.
Concerns about pill burdens and opportunities presented by patent expiries will drive the co-formulation of aspirin with antihypertensives and lipid-lowering agents. The vast costs associated with elderly populations and rampant atherosclerotic disease will also force more cost sensitive solutions (e.g. generics) and improved rationing strategies (better risk calculation algorithms and sub-stratification techniques) on health services. It may eventually dawn on health services that prevention of atherosclerosis is simple, cheap, effective and far less costly than attempts at curing the disease once established.