Summary
The proprotein convertase subtilisin/kexin type 9 (PCSK9) regulates cholesterol metabolism mainly by targeting the low-density lipoprotein receptor (LDLR) for degradation in the liver. Gain-of-function mutations in PCSK9 are one of the genetic causes of autosomal dominant hypercholesterolaemia. Conversely, loss-of-function mutations are associated with lower concentrations of LDL cholesterol (LDL-C) and reduced coronary heart disease. As these loss-of-function mutations are not associated with apparent deleterious effects, PCSK9 inhibition is an attractive new strategy for lowering LDL-C concentration. Among the various approaches to PCSK9 inhibition, human data are only available for inhibition of PCSK9 binding to LDLR by monoclonal antibodies. In phase II studies, the two most advanced monoclonal antibodies in development (alirocumab and evolocumab) decreased atherogenic lipoproteins very effectively and were well tolerated. A dramatic decrease in LDL-C up to 70% can be obtained with the most efficacious doses. Efficacy has been evaluated so far in addition to statins in hypercholesterolaemic patients with or without familial hypercholesterolaemia, in patients with intolerance to statin therapy and in monotherapy. Large phase III programmes are ongoing to evaluate the long-term efficacy and safety of these very promising new agents.
Résumé
PCSK9 ou proprotein convertase subtilisin/kexin type 9 est une protéine clé dans la régulation du métabolisme du cholestérol qui agit principalement en augmentant la dégradation du récepteur des lipoprotéines de basse densité (LDLR) dans le foie. Des mutations « gain-de-fonction » de PCSK9 sont l’une des causes génétiques de l’hypercholestérolémie autosomique dominante. À l’opposé, des mutations « perte-de-fonction » ont été associées avec des taux bas de LDL-cholestérol (LDL-C) et une réduction du risque de maladie coronarienne. Comme ces mutations « perte-de-fonction » n’induisent pas d’effets délétères, l’inhibition de PCSK9 est une nouvelle stratégie intéressante pour abaisser les taux plasmatiques de LDL-C. Parmi les diverses approches pour inhiber PCSK9, des données humaines sont seulement disponibles pour l’instant avec des anticorps monoclonaux qui inhibent la liaison de PCSK9 aux LDLR. Dans les études de phase II, les deux anticorps monoclonaux les plus avancés dans le développement (alirocumab et évolocumab) se sont révélés très efficaces pour diminuer les lipoprotéines athérogènes et ont été bien tolérés. Une diminution majeure du LDL-C jusqu’à 70 % a été observée avec les doses les plus efficaces. L’efficacité a été évaluée jusqu’alors en addition aux statines chez des patients hypercholestérolémiques avec ou sans hypercholestérolémie familiale, chez des patients avec intolérance à un traitement par statine, et en monothérapie. De larges programmes de phase III sont en cours pour évaluer l’efficacité et la sécurité d’emploi à long terme de ces nouveaux agents très prometteurs.
Background
High concentrations of low-density lipoprotein cholesterol (LDL-C) have consistently been associated with an increased risk of atherosclerotic cardiovascular disease, particularly coronary heart disease (CHD). Low-density lipoprotein (LDL) particles are removed from the circulation mainly by hepatic uptake via the LDL receptor (LDLR). LDL binds to the LDLR and the LDL/LDLR complex is internalized into clathrin-coated vesicles by endocytosis. Then, LDL is separated from its receptor in the endosomes and the LDLR is recycled for reuse. At the same time, LDL is degraded ( Fig. 1 a) .
Both environmental and genetic factors regulate plasma concentrations of LDL-C. Among genetic causes, autosomal dominant hypercholesterolaemia (ADH) is associated with elevated LDL-C concentration and premature cardiovascular disease. In the majority of the cases, familial hypercholesterolaemia (FH) is related to mutations in the LDLR itself. A second and less frequent form of FH is caused by mutations in the ligand-binding domain of apolipoprotein (apo) B100, the protein component of LDL that interacts with the LDLR . The identification of a third gene associated with FH, encoding proprotein convertase subtilisin/kexin type 9 (PCSK9), has generated intensive research into PCSK9, making this protein a key regulator for LDLR activity and an attractive target for the treatment of hypercholesterolaemia .
The discovery of PSCK9
In 2003, Seidah et al. identified the ninth member of the proprotein convertase family, PCSK9 . In the same year, the involvement of PCSK9 in regulating cholesterol metabolism became evident, with the identification of two gain-of-function (GOF) mutations in PCSK9, in two French families with a clinical diagnosis of ADH and no detectable mutations in LDLR or apoB100 genes . Since this first report, several other GOF mutations have been reported , associated with mild to severe hypercholesterolaemia and an increased risk of CHD . However, GOF mutations in PCSK9 are relatively rare and account for a small percentage of patients with ADH .
Conversely, the genetic evidence suggesting a potential role for PCSK9 inhibition in decreasing LDL-C concentration came from the identification of loss-of-function (LOF) mutations and common polymorphisms associated with lower LDL-C concentrations. The first LOF mutations were described in 2005 and the effect of lifelong reductions in LDL-C induced by these LOF mutations was examined in the atherosclerosis risk in communities study : the LOF mutations Y142X and C679X in African Americans were associated with a 28% reduction in LDL-C and an 88% reduction in the risk of CHD, whereas the R46L mutation in Caucasians was associated with a 15% reduction in LDL-C and a 47% reduction in the risk of CHD .
Numerous other LOF mutations or polymorphisms associated with decreased LDL-C concentrations have been identified . The association between the R46L mutation and the risk of CHD has been extensively evaluated in three independent Danish studies . In meta-analyses, R46L carriers had a 12% reduction in LDL-C and a 28% reduction in CHD risk . The fact that CHD risk reduction was considerably larger than predicted with similar LDL-C reductions in statin trials could be explained by the effect of long-term exposure to lower LDL-C beginning early in life. This is also in agreement with the results of a Mendelian randomization analysis, in which long-term exposure to lower LDL-C was associated with a three-fold greater reduction in CHD risk than that observed during treatment with a statin started later in life .
Structure and biosynthesis of PCSK9
The human PCSK9 gene located on chromosome 1p32.3 encodes a 692-amino acid inactive glycoprotein. PCSK9 is expressed in several organs, particularly the liver and also the intestine and the kidney . The 692-amino acid preproPCSK9 undergoes signal peptidase cleavage (domain 1–30), then autocatalytic cleavage in the endoplasmic reticulum into two products: the prodomain and the mature PCSK9 containing the catalytic domain and the C-terminal domain. Cleavage of the prodomain is required for the maturation and secretion of PCSK9. Indeed, a recent LOF mutation due to an amino acid substitution within the cleavage site prevents autocatalytic processing and, by inhibiting PCSK9 secretion, is associated with a 48% reduction in LDL-C concentration .
After cleavage, the prodomain or prosegment remains associated by hydrogen bonds with the mature active form and the protein is finally secreted as an inactive dimer complex ( Fig. 1 b). Recent reviews have examined the role of this complex in preventing the access of other potential substrates to the catalytic domain of PCSK9 . The activity of PCSK9 in promoting LDLR degradation seems independent of its catalytic activity. The catalytic domain of mature PCSK9 binds to the first epidermal growth factor-like repeat A domain of the LDLR, while the C-terminal domain binds to cell surface proteins, including Annexin A 2 .
Functions of PCSK9
Role in the regulation of LDL-C concentration
The major function of PCSK9 is the degradation of the LDLR by complex mechanisms : PCSK9 directly interacts with the LDLR both within the cell and at the surface of the plasma membrane . However, evidence indicates that PCSK9 acts on the LDLR primarily as a secreted factor and promotes the reduction of LDLR protein concentrations, mainly in the liver. LDLR protein concentrations are increased in the liver of PCSK9 knockout mice .
Secreted PCSK9 binds to the LDLR in a complex with its prosegment and is subsequently internalized together with the LDLR. The binding of PCSK9 to LDLR induces modification of LDLR conformation, avoiding normal recycling of LDLR to the plasma membrane and enhancing the LDLR lysosomal degradation ( Fig. 1 b).
As a result, LDLR represents the main route of elimination of PCSK9 . However, the mature secreted PCSK9 can be inactivated through cleavage by other proprotein convertases, particularly furin. The mature active form and the inactive form of PCSK9 circulate in the bloodstream.
Other functions of PCSK9
Although PCSK9 is a key player in cholesterol homeostasis through the regulation of LDLR concentrations, data suggest a role in triglyceride metabolism and triglyceride accumulation in visceral adipose tissue . The function of PCSK9 in the intestine is not well known. It has been recently reported that PCSK9 can enhance chylomicron secretion and participate in the control of enterocyte cholesterol balance .
Beyond effects on lipid metabolism, animal data also suggest a role for PCSK9 in glucose homeostasis , liver regeneration and susceptibility to hepatitis C virus infection . Although unexpected adverse effects cannot be excluded during PCSK9 inhibition, genetic variants of PCSK9 have given reassuring information. One of the most striking examples is that of a woman who carries two mutations in PCSK9 with no detectable circulating PCSK9, an LDL-C concentration of 14 mg/dL and normal hepatic, neuronal and renal functions .
Recently, it has been reported that absence of PCSK9 can be protective against melanoma invasion in mouse liver , suggesting that a PCSK9 inhibitor may be also useful in therapies against cancer metastasis. However, there is a large need for human data and, until now, PCSK9 inhibitors have been developed to treat hypercholesterolaemia and prevent atherosclerosis.
PCSK9 and atherosclerosis in animal models
In mice lacking PCSK9, the accumulation of cholesteryl esters in aortas was markedly reduced. By comparison, overexpression of PCSK9 induced an excess of atherosclerosis . But in LDLR-deficient mice lacking or overexpressing PCSK9, no significant differences were observed in cholesteryl ester accumulation and plaque size, strongly suggesting that the process by which PCSK9 enhances atherosclerosis is primarily mediated through its action on the LDLR .
Cloned minipigs created by transposition of a human PCSK9 GOF mutant – a model for FH – had a significant increase in aortic atherosclerosis compared with wild-type minipigs .
Regulation of PCSK9 plasma concentrations
PCSK9 gene expression is mainly modulated by intracellular cholesterol concentrations and consequent activation of the transcription factor sterol-responsive element-binding protein 2 (SREBP2) , similarly to other genes involved in cholesterol homeostasis, such as LDLR. This concomitant regulation of both PCSK9 and LDLR by cholesterol via SREBP2 helps to explain the ‘paradoxical effect’ of statin therapy : statins not only upregulate expression of the LDLR, but also expression of PCSK9, potentially limiting the pharmacological effect of reducing LDL-C concentration ( Fig. 1 c). Statin administration to PCSK9 knockout mice enhanced LDL clearance from plasma . These data support PCSK9 inhibition as a very attractive target for lowering LDL-C and enhancing the efficacy of statin treatment.
Strategies for PCSK9 inhibition
Several therapeutic approaches to the inhibition of PCSK9 have been proposed , including: inhibition of PCSK9 synthesis by gene silencing agents, such as antisense oligonucleotides or small interfering RNA (siRNA); inhibition of PCSK9 binding to LDLR by monoclonal antibodies (mAbs), small peptides or adnectins; and inhibition of PCSK9 autocatalytic processing by small molecule inhibitors. These strategies, targeting either extracellular or intracellular PCSK9, have been extensively described in recent reviews .
Preclinical studies on inhibition of PCSK9 synthesis by antisense oligonucleotides were promising, but the development of two antisense oligonucleotides by BMS/ISIS (BMS-84421) and Santaris Pharma (SPC5001) was stopped in phase I . siRNA is another approach : in a phase I trial of ALN-PCS – an siRNA developed by Alnylam Pharmaceuticals – a dose-dependent reduction in LDL-C was observed, with a 41% reduction at day 4 with the highest dose, associated with a 68% reduction in plasma PCSK9 concentrations . Inhibition of PCSK9 binding to LDLR by small peptide inhibitors such as SX-PCK9 (Serometrix, East Syracuse, NY, USA) or adnectins such as BMS-962476 (BMS/Adnexus, Waltham, MA, USA) are in preclinical development or phase I . On the basis of the discovery of an LOF mutation in the autocatalytic cleavage site of PCSK9 , inhibition of PCSK9 autocatalytic processing is the approach chosen by Cadila Healthcare and Shifa Biomedical , with molecules in preclinical development phase. Finally, mAbs are currently the most advanced approach in terms of clinical development, with published phase I and phase II human trials ( Fig. 1 c).
Efficacy of monoclonal antibodies targeting PCSK9
Several mAbs targeting PCSK9 have been tested in preclinical studies to assess their disruption of the PCSK9-LDLR interaction or inhibition of PCSK9 internalization . Human data are available for three of these mAbs: SAR236553/REGN727 (alirocumab) and AMG145 (evolocumab), two fully human mAbs developed by Sanofi/Regeneron and Amgen, respectively; and RN316/PF04950615, a humanized mAb developed by Pfizer/Rinat.
Alirocumab (SAR236553/REGN727)
Three phase I studies of alirocumab have been performed, two in healthy volunteers and one in patients with hypercholesterolaemia . In the two single ascending-dose studies, alirocumab reduced LDL-C in a dose-dependent manner by 28–65% when given intravenously (0.3–12.0 mg/kg) and by 32–46% when given subcutaneously (50–250 mg). In the third placebo-controlled multiple-dose trial, alirocumab was administered subcutaneously at doses of 50, 100 or 150 mg in adults receiving atorvastatin with LDL-C > 100 mg/dL and at the 150 mg dose in adults on diet alone with LDL-C > 130 mg/dL. Alirocumab significantly reduced LDL-C by 38–65% in patients taking atorvastatin and by 57% in patients not taking atorvastatin. Alirocumab induced a maximum lowering of LDL-C within 2 weeks. It seems that the duration of action is longer in subjects who are not treated with atorvastatin, suggesting that the statin-stimulated production of PCSK9 might affect the duration of action of therapeutic mAbs.
The efficacy of alirocumab administered subcutaneously was examined in three phase II randomized double-blind placebo-controlled trials . The efficacy on atherogenic lipid variables observed with the most efficacious dose is indicated in Table 1 .
