Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
Review
Paroxysmal nocturnal hemoglobinuria and the age of therapeutic complement inhibition Expert Rev. Clin. Immunol. 9(11), 1113–1124 (2013)
Juan Carlos Varela* and Robert A Brodsky Department of Medicine, The Johns Hopkins School of Medicine, Division of Hematology, 720 Rutland Ave., Ross Research Building, Room 1025, Baltimore, MD, 21205, USA *Author for correspondence: Tel.: +1 410 502 2546 Fax: +1 410 955 0185 [email protected]
Paroxysmal nocturnal hemoglobinuria (PNH) is a rare disease of hematopoietic stem cells due to a mutation in the PIG-A gene leading to a deficiency of GPI-anchored proteins. Lack of two specific GPI-anchored proteins, CD55 and CD59, leads to uncontrolled complement activation that result in both intravascular and extravascular hemolysis. Free hemoglobin leads to nitric oxide depletion that mediates the pathophysiology of some of the common clinical signs of PNH. Clinical symptoms of PNH include evidence of hemolytic anemia, bone marrow failure, smooth muscle dystonias and thromboses. Treatment options for patients with PNH include bone marrow transplantation, a therapy associated with high morbidity and mortality, or treatment with the complement inhibitor eculizumab. Eculizumab is a first-in-class anti-complement drug that in PNH has been shown to block complement-mediated hemolysis, reduce transfusion dependency, reduce thromboembolic complications and improve the quality of life (QoL) of patients. KEYWORDS: complement inhibition • eculizumab • paroxysmal nocturnal hemoglobinuria
Paroxysmal nocturnal hemoglobinuria (PNH) is a rare clonal disease of hematopoietic stem cells. Patients with PNH can present with bone marrow failure, hemolytic anemia, smooth muscle dystonias and thrombosis [1–3]. The natural history of this disease is highly variable and the estimated median survival without specific therapy is approximately 10–20 years [4–6]. PNH results from somatic mutations in the phosphatidylinositol glycan class A (PIG-A) gene, which is required for the synthesis of glycophosphatidylinositol (GPI) anchors [7,8]. As a result of these mutations, the affected stem cell and all of its progeny have a deficiency or absence of GPIanchored proteins. Central to the pathophysiology of PNH is the absence of two specific GPIanchored proteins, CD55 and CD59 [9,10]. These two proteins are membrane-bound complement inhibitory proteins which function to protect cells from complement attack. Specifically, CD55 regulates the early phase of the complement cascade by inhibiting C3 convertases while CD59 regulates the terminal phase of the complement cascade by inhibiting the formation of
www.expert-reviews.com
10.1586/1744666X.2013.842896
the membrane attack complex [11]. The absence of these complement regulators on the surface of erythrocytes makes these cells susceptible to complement attack leading the clinical signs and symptoms observed in PNH. Complement activation on the surface of the affected cells leads to opsonization with complement fragments and formation of the membrane attack complex. These processes lead to intravascular and extravascular hemolysis [12–14]. Intravascular hemolysis is at the center of the pathophysiology of PNH, while extravascular hemolysis was not recognized until effective therapies to block intravascular hemolysis were developed. Intravascular hemolysis leads to release of free hemoglobin, which in turn scavenges and depletes nitric oxide. It is this depletion of nitric oxide that contributes to the clinical manifestation of PNH including thrombosis, smooth muscle dystonias (e.g., esophageal spasms, male erectile dysfunction) and renal insufficiency [5,6,15]. The severity of symptoms is variable among patients diagnosed with PNH and not all patients require treatment. The only known curative treatment for PNH is bone
Ó 2013 Informa UK Ltd
ISSN 1744-666X
1113
Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
Review
Varela & Brodsky
marrow transplantation; however, this therapy is associated with high morbidity and mortality [16–19]. Until the early 2000’s, the only other available treatment for PNH was management of individual symptoms (i.e., transfusions for anemia, anticoagulation for thrombosis). In 2002, eculizumab, a monoclonal antibody against the complement protein C5, was first used successfully in the treatment of patients with PNH [20]. C5 is the central protein in the complement cascade and its blockade prevents complement attack via the MAC regardless of the pathway of activation. Significant clinical improvements were realized by blocking the terminal pathway of complement with eculizumab in patients with PNH. In the initial pilot study of 11 patients with PNH, treatment with eculizumab reduced intravascular hemolysis, hemoglobinuria, reduced the need for transfusion and showed a positive effect on the quality of life (QoL) in patients with PNH. Subsequently, two Phase III trials (TRIUMPH and SHEPHERD), confirmed the results obtained in the initial pilot study [21,22]. These findings led to the approval of eculizumab for the treatment of patients with PNH by the US FDA and the European Medicines Agency (EMA) in 2007. This article will focus on the pathophysiology of PNH, rationale for targeting the complement system in PNH, the development and clinical use of eculizumab as treatment for PNH and the prospects for further therapies for the treatment of PNH. PIG-A mutation & deficiency of GPI-anchored proteins
The central defect in stem cells and their progeny in patients with PNH is a somatic mutation in the gene encoding for PIG-A. PIG-A is an enzyme involved in the biosynthesis of GPI-anchors, and these anchors are required for the attachment of dozens of proteins to the plasma membrane of cells. The synthesis of GPI anchors is a complex biological process that involves several biochemical reactions and the interaction of more than 20 gene products [23]. PIG-A, in concert with six other enzymes, catalyzes the first step in GPI-anchored biosynthesis. Specifically, it catalyzes the transfer of Nacetylglucosamine (GlcNAc) to phosphatidylinositol (PI) forming GlcNAc-PI. From this crucial step, the synthesis of GPI anchors continues in the endoplasmic reticulum and the completed product is transferred en bloc to the carboxyl terminus of proteins that have a GPI-attachment signal peptide. Early studies of PNH blood cells revealed an absence of specific cell surface proteins. Some of the proteins identified as missing at that time were leukocyte alkaline phosphatase, erythrocyte acetylcholinesterase and later decay accelerating factor (CD55). While absence of the first two proteins did not explain the pathophysiology of PNH, absence of the latter hinted at the mechanism of hemolysis. At the time, it was not known that these proteins were attached to the cell membrane by GPI anchors and it was not until this was determined that the link between PNH and absence of GPI-anchored proteins was established [24,25]. Subsequent studies using cell lines derived from PNH patients determined that the absence of GPI-anchored proteins on the cell surface was due to a defect in the biosynthesis of GPI anchors. Specifically, it was found that cells from PNH patients were unable to carry out the first 1114
step in GPI anchor biosynthesis and were unable to synthesize GlcNAc-PI [26]. Further investigations provided a convincing link between PNH and PIG-A as they revealed that the observed defect in GPI anchor biosynthesis was due to a somatic mutation of the PIG-A gene [27,28]. PNH stem cells & clonal expansion
Deficiency of GPI-anchored proteins occurs in all blood lineages in patients with PNH. The affected cells are clonal and have common PIG-A mutations proving that these mutations arises multipotent hematopoietic stem cells [29,30]. PIG-A mutations have also been described in patients with aplastic anemia (AA), myelodysplastic syndrome (MDS) and in normal healthy controls [30–33]. Similar to PNH, the PIG-A mutations found in patients with AA occur in multipotent hematopoietic stem cells. While AA patients do not show the clinical signs of PNH (hemolysis, smooth muscle dystonias and thrombosis) early in their disease, many AA patients with a PNH clone eventually develop classical PNH. In contrast, MDS patients do not develop PNH. This observation can be explained by the finding that the PIG-A mutations observed in MDS patients and normal controls are transient and have been found to occur in more differentiated colony-forming cells. These differentiated colonyforming cells do not have the self-renewal capacity of multipotent hematopoietic stem cells and are quickly exhausted [30,33]. The mechanisms leading to the clonal expansion and dominance of PNH stem cells remain a topic of continued investigation. The leading hypothesis is that PNH stem cells have a conditional advantage in the setting of an autoimmune attack (e.g., aplastic anemia) [34]. The precise mechanism remains unclear but it may be related, in part, to the absence of GPIanchored ULBP ligands on the surface PNH cells; the receptor for these ULBP ligands in NKG2D, an activating receptor on natural killer cells and CD8+ T cells [35,36]. The role of the complement system in PNH
The complement system is one of the main components of the innate immune system and it is a powerful and effective mechanism that protects the host from invading pathogens. It is made up of more than 35 serum and membrane-bound proteins that interact in a series of tightly regulated, proteolytic steps leading to the induction of a number of biological processes with the main objective of protecting the host from pathogens such as bacteria, viruses, virus-infected cells, parasites and tumor cells (FIGURE 1). In addition, the complement system plays an important role in the disposal of waste (i.e., clearance of immune complexes and apoptotic cells) and it acts as a bridge between innate and adaptive immunity [11,37]. Activation of complement can be achieved via three different pathways: the classical, alternative and lectin. Each of these pathways is initiated by different stimuli, but all lead to the activation of the central complement proteins, C3 and C5. Subsequent steps in the complement cascade result in the activation of a number of complement-dependent mechanisms of high importance for both the innate and adaptive immune responses. Activation of Expert Rev. Clin. Immunol. 9(11), (2013)
Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
PNH & the age of therapeutic complement inhibition
Review
the complement system is tightly reguClassical pathway Alternative pathway lated via soluble and membrane bound and 'tickover' lectin pathway proteins. There are two main points of regulation of the complement cascade, CD55 control of activation of the C3 convertases and control of the formaC3 convertase C3 convertase C3 tion of the membrane attack complex (C3bBb) (C4b2a) (MAC). In erythrocytes, these regulatory functions are carried out by the membrane bound proteins, CD55 and CD59. Amplification loop CD55 inhibits the formation of the C3b C3 convertases as well as accelerates their decay in all pathways of complement activation [38]. CD59 inhibits the terminal pathway and directly prevents the formaC3b Eculizumab C5 tion of the MAC [10]. Both CD55 and CD59 are bound to the cell surface via GPI anchors and are absent in PNH MAC C5a C5b erythrocytes rendering these cell susceptible to complement attack [9,39]. Complement activation on the surface of PNH Extravascular CD59 erythrocytes leads to intravascular and hemolysis extravascular hemolysis by two different mechanisms (FIGURE 1). Central to these mechanisms is the alternative pathway of complement activation. In this pathway, C3 protein spontaneously hydrolyzes and Free hemoglobin leads to the formation of C3 convertase Hemoglobin (this process is also known as ‘tickover’). The mechanism of intravascular hemolysis begins with the increased activity of C3 convertases on the surface of PNH erythrocytes as a result of the lack of CD55. This leads to activation of C3, Intravascular hemolysis C5 and the terminal pathway of complement culminating in the formation of the Figure 1. The complement system, paroxysmal nocturnal hemoglobinuria and MAC. Under normal conditions, formaeculizumab. The complement system is activated via the classical, lectin or alternative tion of the MAC is under the regulation pathways. C3 is activated via C3 convertases. This step is regulated by the action of of CD59. The absence of CD59 on CD55, a GPI-anchored protein. Subsequently, C5 is cleaved into C5a and C5b. C5a mediates a number of biological processes while C5b begins the terminal pathway of complePNH erythrocytes leads to uncontrolled ment and the assembly of the MAC. Deposition of MAC on the surface of cells leads to formation of the MAC resulting in cell lysis. The formation of the MAC is regulated by the action of CD59, another GPIcomplement – mediated intravascular anchored protein. Paroxysmal nocturnal hemoglobinuria (PNH) cells have a deficiency in hemolysis. The mechanism of extravascuGPI-anchored proteins on their cell surface. Absence of CD55 and CD59 leads to unconlar hemolysis begins with increased opsotrolled complement activation on the surface of PNH cells. Deficiency of CD59 increases MAC formation and induces intravascular hemolysis, which is central to the pathophysiolnization of PNH erythrocytes by ogy of PNH. Deficiency of CD55 leads to increased C3 convertase activity and C3dcomplement fragments (mostly C3d). associated extravascular hemolysis. Eculizumab is a humanized monoclonal antibody that This is the result of the lack of CD55. targets C5. By preventing C5 activation, eculizumab prevents the formation of the MAC Opsonized erythrocytes are cleared and leading to a significant reduction in intravascular hemolysis of PNH cells. Use of eculizudestroyed by cells of the reticulomab unmasks a low-level C3-d mediated extravascular hemolysis. MAC: Membrane attack complex. endothelial system [40]. While both CD55 and CD59 are missing on the surface of PNH erythrocytes, it is the absence of CD59 leading to hemolysis, data has demonstrated that unregulated complement unregulated activation of the terminal pathway of complement activation also affects platelets, leukocytes and endothelial cells and the resulting intravascular hemolysis that drives the patho- in patients with PNH. It has been shown that platelets lacking physiology of the disease. In addition to precipitating CD59 produce more procoagulant vesicles [41,42]. Complement www.expert-reviews.com
1115
Review
Varela & Brodsky
activation on macrophages and monocytes, leads to cell injury and release of tissue factor [43]. Most recently, studies have shown evidence of endothelial cell dysfunction in patients with PNH [44,45].
Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
Complement-mediated hemolysis & nitric oxide scavenging
Complement-mediated intravascular hemolysis leads to the release of free hemoglobin into the plasma of PNH patients. Free hemoglobin is normally cleared by haptoglobin, CD163 and hemopexin [12]. The extensive hemolysis seen in PNH overwhelms the clearing mechanisms and leads to accumulation of high levels of free hemoglobin in the plasma. The main consequence of this process is the depletion of nitric oxide (NO). Free hemoglobin is a potent NO scavenger as the two undergo a fast and irreversible reaction that results in the production of nitrate (NO3) and methemoglobin. Normally, NO is synthesized by endothelial cells and functions to maintain smooth muscle relaxation and inhibit platelet activation and aggregation. Additionally, erythrocyte arginase is released as a result of hemolysis leading to depletion of L-arginase, a precursor of NO, resulting in decreased production of NO. The absence of NO as a result of scavenging by free hemoglobin and decreased production due to low L-arginase levels leads to deregulation of muscle tone, endothelial dysfunction and inappropriate activation of platelets (FIGURE 2). In turn, these processes lead to smooth muscle dystonias and propensity for thrombosis. Diagnosis & clinical aspects of PNH
PNH is diagnosed based on clinical symptoms and laboratory studies. Clinical signs can include symptoms from hemolytic anemia, thromboses, hemoglobinuria, smooth muscle dystonias (e.g., esophageal spasms, male erectile dysfunction), abdominal pain, arterial/pulmonary hypertension and renal insufficiency [1,4,5,15]. Laboratory studies used for the diagnosis of PNH include basic hematological tests including complete blood counts, reticulocyte count, lactate dehydrogenase (LDH), haptoglobin levels, flow cytometric evaluation of GPI anchored proteins [46–48] and sometimes bone marrow biopsy. PNH can be divided into three subcategories, classical PNH, PNH in the setting of another specified bone marrow disorder and subclinical PNH. Patients with classical PNH have signs and symptoms of intravascular hemolysis (i.e., increased reticulocyte count, increased serum LDH) and have no evidence of another bone marrow disorder. While not required for the diagnosis of PNH, bone marrow evaluation in these patients reveals a hypercellular bone marrow with erythroid hyperplasia but without karyotypic abnormalities [2,49]. Patients with PNH in the setting of another specified bone marrow disorder have clinical and laboratory evidence of hemolysis and also have another defined bone marrow abnormality (presently or historically). Patients with subclinical PNH have no evidence of hemolysis but have a small clone of PNH cells. These patients can progress to clinically significant PNH, as it is the case with patients 1116
with acquired AA. At diagnosis, most patient with AA have 60% also appear to be at higher risk for thrombosis [5,15]. The mechanisms leading to thrombosis in PNH are not fully elucidated, but several processes have been associated with the prothrombotic state. These processes include NO scavenging by free hemoglobin leading to deregulation of platelet activation and aggregation, production of C5a during complement activation leading to upregulation of tissue factor, generation of PNH plateletderived microvesicles with phosphatidylserine, a known in vitro procoagulant [51,42], release of tissue factor from macrophages and monocytes damaged by complement [43] and endothelial cell dysfunction leading to an increased risk of thrombosis [45]. Treatment of PNH before the development of eculizumab
Treatment of PNH is reserved for those with significant disease manifestations such as disabling fatigue, thromboses, transfusion dependence, or renal insufficiency [2]. Bone marrow transplantation is the only known curative treatment for PNH. Different studies have reported a 2-year survival of 56% [52] and 10-year survival of 57% [18]. More recently, nonmyeloablative conditioning regimens have been used with both HLA-identical [53] and HLA-haploidentical donors [19]. Currently, bone marrow transplantation is only recommended for patients with severe manifestations (life-threatening cytopenias) who do not respond to drug therapy and in countries where eculizumab therapy is not available [54]. Most recently a study of 211 patients recommended against the use of allogeneic stem cell transplantation in patients with life-threatening thromboembolism given the availability of eculizumab [55]. Prior to the development of eculizumab and besides bone marrow transplantation, the treatment of PNH was focused on supportive therapy. For the ongoing hemolytic anemia, patients received red blood cell transfusions, iron and folic acid supplementation [56]. Corticosteroids were successful at decreasing complement-mediated hemolysis in some patients, but their toxicity and lack of reliable effect limit their use. The use of prophylactic anticoagulation in patients with PNH remains a topic of controversy. While some groups have suggested Expert Rev. Clin. Immunol. 9(11), (2013)
PNH & the age of therapeutic complement inhibition
Review
Free hemoglobin
Fe2+
Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
NO
O2
NO NO Complement-mediated intravascular hemolysis
O2 O2
NO
O2
Nitric oxide scavenging
+
NO3-
Fe3+ Nitric oxide depletion
Impaired regulation of smooth muscle tone
Platelet activation and aggregation
Smooth muscle dystonias (abdominal pain, esophageal reflux, erectile dysfunction
Thrombosis
Figure 2. Complement-mediated intravascular hemolysis leads to nitric oxide scavenging and depletion in paroxysmal nocturnal hemoglobinuria. Complement-mediated intravascular lysis of paroxysmal nocturnal hemoglobinuria erythrocytes leads to leakage of free hemoglobin into the plasma. In turn, free hemoglobin scavenges and depletes NO. Depletion of NO leads to dysregulation of smooth muscle tone leading to the dystonias associated with paroxysmal nocturnal hemoglobinuria (i.e., esophageal spasms, erectile dysfunction). In addition, depletion of NO leads to platelet activation and aggregation, this is believed to contribute to the increased risk of thrombosis. NO: Nitric oxide.
primary prophylaxis with warfarin in PNH patients with a granulocyte clone of more than 50% and no contraindications to anticoagulation [57], other groups suggests that guidelines for primary prophylaxis cannot be made based on the available data [49]. Most recently, a study from the South Korean registry, concluded that patients with LDH levels >1.5 ULN at diagnosis were at a significantly higher risk of thromboembolism and that the combination of increased LDH (>1.5 ULN) with the clinical symptoms of abdominal pain, chest pain, dyspnea or hemoglobinuria was associated with a greater increased risk for thromboembolism [58]. www.expert-reviews.com
Eculizumab: the first anti-complement drug
The complement system has been implicated in the pathogenesis of a number of human diseases. Searching for therapeutic agents targeting the complement system has been a subject of much discussion and research. An ideal anti-complement therapeutic would target the complement cascade at the point of convergence of all of the activating pathways and before the cleavage of C5. Therefore, developing therapeutics targeting the C5 protein would be the most rational approach. In the 1990’s, a number of murine anti-human C5 monoclonal antibodies were screened with the goal of finding an antibody that 1117
Review
Varela & Brodsky
Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
Lactate dehydrogenase (U/l)
3000 2500 2000 Eculizumab (n = 43) 1500
Placebo (n = 44)
have an important role in treatment of the disease. It was expected that blockade of C5 with eculizumab would lead to inhibition of the terminal pathway of complement, prevention of MAC assembly and inhibition of complementmediated intravascular hemolysis (FIGURE 1)
1000
Eculizumab for PNH 500
Initial Phase II pilot study
The first study of eculizumab in patients with PNH was started in May of 0 10 20 30 40 50 2002 [20]. It was a 3-month, open-label, Time, weeks Phase II pilot study which enrolled eleven PNH patients (six men and five women), Figure 3. Treatment with eculizumab leads to a significant reduction in complement-mediated intravascular hemolysis. Mean levels of lactate dehydrogenwho had a detectable GPI-deficient hemaase (LDH) reflect the degree of hemolysis. The dashed line represents the upper limit of topoietic clone and had received at least the normal range for LDH (103–230 U/l). In patients treated with eculizumab, the level of four red cell transfusions in the preceding LDH was rapidly and significantly reduced to levels just above the normal range. In the 12 months. The study investigated the placebo group, the mean LDH level remained highly elevated. The arrow represents the effect of eculizumab on complementtime of transition of placebo-treated patients to treatment of eculizumab under the extension study following the initial TRIUMPH. Initiation of eculizumab treatment in mediated hemolysis as determined by placebo-treated patients led to a significant decreased in LDH to near normal levels. LDH, haptoglobin, bilirubin, hemoglobin, Reproduced with permission from [59]. reticulocyte counts, hemoglobinuria and need for transfusions. In addition, it measwould both inhibit the generation of C5a and the formation of ured the effect on the proportion of GPI-deficient cells as well the MAC. One antibody, m5G1.1, was selected out of more as any effect on the QoL of the study participants. Patients 30,000 screened [59]. Subsequently, this antibody was modified for were given 600 mg of eculizumab iv. weekly for 4 weeks, foluse as a human therapeutic. First, in order to minimize immuno- lowed by a 900 mg dose on week 5 and then by a 900 mg genicity, m5G1.1 was humanized by grafting its complementarity- dose every other week through week 12. In this pilot study, determining regions in human heavy and light chain antibody eculizumab was shown to significantly decrease complementframeworks. In addition, in order to eliminate antibody-mediated mediated hemolysis as evidenced by a significant decreased in effector functions (i.e., activation of complement, initiation of Fc LDH levels (FIGURE 3), transfusion requirements, hemoglobinureceptor-mediated pathways), the heavy chain constant region of ria and number of paroxysms. Interestingly, the LDH levels m5G1.1 was replaced with components of both human IgG2 and decreased to mean above the highest limit of normal, indicatIgG4 constant regions [59]. These IgG subtypes were chosen ing some persistent low-level hemolysis. The QoL of particibecause IgG2 is known to not bind Fc receptors and IgG4 does pants was also evaluated and found to be significantly not activate complement. The resulting antibody was the first improved. Importantly, this pilot study also investigated the anti-complement drug, eculizumab, a humanized monoclonal safety, pharmacokinetics and immunogenicity of eculizumab. antibody targeting the complement protein C5, which inhibited There were no treatment-related deaths or thrombotic events. the generation of C5a, prevented the formation of the MAC, was One of the main concerns at the time was the increased risk not immunogenic and did not activate complement or Fc for infection, especially by encapsulated organisms. All study receptor-mediated pathways. participants were vaccinated against Neisseria meningitidis Preclinical in vivo studies of eculizumab were carried out in prior to receiving eculizumab and no infectious complications mouse models of arthritis and lupus. Of note, given the were reported. The study also showed that therapeutic levels humanized nature of eculizumab, a mouse anti-C5 antibody of eculizumab were reached and maintained throughout the that demonstrated similar activity to eculizumab was used study period. No antibodies against eculizumab were detected instead. In both models, blockade of the complement cascade in any of the participants. Following the initial pilot study, at C5 led to amelioration of disease and prolonged survival of all participants continued treatment in a 52-week extension mice [60]. The first human Phase I trials of eculizumab were study [61]. The yearlong study confirmed all the findings of performed in patients with rheumatoid arthritis, membranous the pilot study. nephritis and systemic lupus erythematosus (SLE). All these studies were Phase I trials aimed at determining the safety and Phase III trials dosing of eculizumab in humans rather than testing its efficacy After the compelling data from the initial pilot study, eculizuagainst disease [59]. While no Phase I trials were carried out in mab received orphan designation for PNH and two Phase III PNH patients, it quickly became apparent that this drug could trials were started in the USA, Europe and Australia. 0
1118
Expert Rev. Clin. Immunol. 9(11), (2013)
Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
PNH & the age of therapeutic complement inhibition
The first study was TRIUMPH; a double-blind, randomized, placebo-controlled, multicenter, Phase III trial [21]. Participants were required to have a PNH type III erythrocyte proportion of at least 10%, LDH levels of at least 1.5-times the upper limit of normal, a platelet count of at least 100,000 per mm3, and to have received at least four transfusions during the previous 12 months. A total of 87 patients (35 men and 52 women) were enrolled in the trial. Patients were randomized into an eculizumab group (n = 43) or a placebo group (n = 44). Patients in the treatment group received 600 mg infusion of eculizumab weekly for 4 weeks, followed by a 900 mg infusion on week 5 and then a 900 mg infusion every other week until week 26. The study examined the effect of eculizumab on hemoglobin levels, transfusion requirements, complement-mediated hemolysis and participant’s QoL. Similar to the pilot the study, TRIUMPH demonstrated the efficacy of eculizumab in the treatment for PNH. It demonstrated a stabilization of hemoglobin levels in 49% of patients in the treatment group compared to 0% in the placebo group (p < 0.001). Transfusion requirements were significantly different in the treatment versus placebo groups. Within 6 months prior to the initiation of the study treatment group patients received 413 units of pack red blood cells versus 417 units for the placebo group. In marked contrast, by week 26 of the study, the treatment group had received 131 units versus 482 units for the placebo group. Hemolysis was significantly decreased in treated patients as evidenced by decreased LDH levels and increased proportions of PNH type III erythrocytes. QoL was also significantly improved in the treatment group compared to the placebo group. There were no treatment related deaths and no serious adverse events were attributed to the treatment. The most common adverse events reported in the eculizumab group were headache and back pain. One patient in each group had low, detectable levels of antibodies against eculizumab, but they were not clinically significant. All study participants were vaccinated against Neisseria meningitides prior to receiving eculizumab and no infectious complications were reported. The second study was SHEPHERD, an open-label, Phase III trial designed to investigate the long-term safety and efficacy of eculizumab by relaxing the inclusion criteria of the TRIUMPH trial [22]. In contrast to the eligibility criteria for TRIUMPH, this trial allowed patients with platelet counts as low as 30,000 per mm 3 and patients were only required to have had at least 1 transfusion in the previous 2 years. Similar to TRIUMPH, participants were required to have a PNH type III erythrocyte proportion of at least 10% and LDH levels of at least 1.5-times the upper limit of normal. A total of 97 patients (48 men and 49 women) were treated in this study. Patient received intravenous infusion of 600 mg of eculizumab weekly for 4 weeks followed by a 900 mg dose a week later and then 900 mg every 14 days for a total of 52 weeks. The study measured adverse events (AEs) possibly related to eculizumab, its effect on hemolysis, fatigue and quality of life of enrolled patients. There was one death www.expert-reviews.com
Review
in the study due to thrombosis, but it was considered unrelated to the study drug. The most common AEs reported were headaches, nasopharyngitis and upper respiratory tract infection which occurred in 52, 32 and 30% of patients, respectively. There were seven serious AEs considered to be possibly related to eculizumab, fever (2), headache (1), abdominal distension (1), viral infection (1), anxiety (1) and renal impairment (1). None of these, however, were considered probably or definitely related to eculizumab. As in previous trials, all patients were immunized against Neisseria meningitides prior to receiving eculizumab. Eighty-nine patients reported at least one infection during the study period. The majority of infections were considered mild to moderate and not related to the study drug. The study also showed that treatment with eculizumab led to a rapid, sustained reduction in hemolysis as determined by significantly decreased LDH levels and increased proportions of PNH type III erythrocytes. The participants also reported a significant decrease in fatigue and improvement in quality of life. Interestingly, 71% of patients in the SHEPHERD study would not have met the inclusion criteria for TRIUMPH; the SHEPHERD trial demonstrated that eculizumab treatment was applicable to broad population of patients with PNH. Additional studies
The main findings from the pilot study, TRIUMPH and SHEPHERD were that eculizumab was an effective and safe treatment for PNH. It reduced the complement-mediated hemolysis, minimized transfusion requirements and improved the QoL of treated patients with reproducible effectiveness and few adverse events. Following the completion of the pilot, TRIUMPH and SHEPHERD studies a total of 187 patients were enrolled in an extension Phase IIIb trial to assess the long-term efficacy safety and efficacy of continued eculizumab treatment [62]. All three original trials followed the same dosing schedule and the patients enrolled in the extension trial continued to receive the maintenance dose of 900 mg of eculizumab every 2 weeks for a total of 66 months. This long-term study was a post hoc analysis and reported that there was a significant reduction in hemolysis, reduction in transfusion dependence, decrease in thromboembolic events and improvement in renal function. Patients treated long-term with eculizumab had a 3-year survival of 97.6% [63]. The AEs were unchanged from the previously reported trials and most commonly included headache, nasopharyngitis and upper respiratory tract infections. Fortypatients reported infectious complications during treatment. Of those, two patients developed meningococcal sepsis and both were treated without complications. Other studies evaluated the effect of eculizumab on other aspects of the pathophysiology of PNH using the same cohort of patients as the three original trials. One study investigated the effect of long-term treatment with eculizumab on thromboembolism (TE) [64]. This study included patients from the 1119
Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
Review
Varela & Brodsky
original pilot trial, TRIUMPH, SHEPHERD and the Phase IIIb extension study. Thromboembolism rate with eculizumab was compared with pretreatment rate in the same patients. The study found a relative reduction of 85% in TE event rate during eculizumab treatment (TE rate before eculizumab = 7.37 events/100 patient years vs, TE rate with eculizumab = 1.07 events/100 patient years; p < 0.001). In patients treated with anticoagulants, there was a relative reduction of 94% with eculizumab (TE event rate was reduced from 10.61–0.62 events/100 patient years, p < 0.001). Another study investigated the effect of eculizumab on NO depletion, dyspnea and measures of pulmonary hypertension [65]. This study was carried out with patients from the TRIUMPH trial only. Treatment with eculizumab significantly reduced NO depletion, dyspnea and decreased the proportion of patients with elevated pro-brain naturetic peptide (proBNP, a marker for pulmonary vascular resistance and right ventricular dysfunction in hemolytic diseases). Treatment with eculizumab beyond the initial trials
Shortly after the completion of TRIUMPH and SHEPHERD, eculizumab became the first drug approved by the FDA and EMA for the treatment of complement-mediated hemolysis in patients with PNH. Most of the published literature regarding the use of eculizumab in patients with PNH has been derived from studies of 187 patients that were enrolled in the clinical trials that lead to FDA approval. To date, few studies have evaluated eculizumab outside the context of a clinical trial. One study reported on 79 patients treated with eculizumab at a single center in the UK [66]. Of the 79 patients, 34 were enrolled in one of the original clinical trials. The patients received eculizumab per the dosing and treatment schedule used in the previous trials. The mean follow-up was 39 months. The study compared survival of PNH patients treated with eculizumab to that of patients with PNH in the 7 years prior to the use of eculizumab and that of the general population. The 5-year survival of patients treated with eculizumab was significantly better than that of patients not treated with eculizumab (95.5 vs 66.8%, respectively, p = 0.0062) and comparable to the survival of the general population (p = 0.46). A second study was a retrospective, single center study that evaluated the response of 30 patients with PNH to treatment with eculizumab in the USA [67]. Of note, out of the 30 patients, five were enrolled on the TRIUMPH or SHEPHERD trial. The patients received eculizumab using the same dosing regimen as the published trials. The patients were categorized based on response criteria. A complete response (CR) was defined as transfusion independence with normal hemoglobin for age and sex, absence of symptoms (i.e., thromboses, smooth muscle dystonias) and a LDH
Review
Paroxysmal nocturnal hemoglobinuria and the age of therapeutic complement inhibition Expert Rev. Clin. Immunol. 9(11), 1113–1124 (2013)
Juan Carlos Varela* and Robert A Brodsky Department of Medicine, The Johns Hopkins School of Medicine, Division of Hematology, 720 Rutland Ave., Ross Research Building, Room 1025, Baltimore, MD, 21205, USA *Author for correspondence: Tel.: +1 410 502 2546 Fax: +1 410 955 0185 [email protected]
Paroxysmal nocturnal hemoglobinuria (PNH) is a rare disease of hematopoietic stem cells due to a mutation in the PIG-A gene leading to a deficiency of GPI-anchored proteins. Lack of two specific GPI-anchored proteins, CD55 and CD59, leads to uncontrolled complement activation that result in both intravascular and extravascular hemolysis. Free hemoglobin leads to nitric oxide depletion that mediates the pathophysiology of some of the common clinical signs of PNH. Clinical symptoms of PNH include evidence of hemolytic anemia, bone marrow failure, smooth muscle dystonias and thromboses. Treatment options for patients with PNH include bone marrow transplantation, a therapy associated with high morbidity and mortality, or treatment with the complement inhibitor eculizumab. Eculizumab is a first-in-class anti-complement drug that in PNH has been shown to block complement-mediated hemolysis, reduce transfusion dependency, reduce thromboembolic complications and improve the quality of life (QoL) of patients. KEYWORDS: complement inhibition • eculizumab • paroxysmal nocturnal hemoglobinuria
Paroxysmal nocturnal hemoglobinuria (PNH) is a rare clonal disease of hematopoietic stem cells. Patients with PNH can present with bone marrow failure, hemolytic anemia, smooth muscle dystonias and thrombosis [1–3]. The natural history of this disease is highly variable and the estimated median survival without specific therapy is approximately 10–20 years [4–6]. PNH results from somatic mutations in the phosphatidylinositol glycan class A (PIG-A) gene, which is required for the synthesis of glycophosphatidylinositol (GPI) anchors [7,8]. As a result of these mutations, the affected stem cell and all of its progeny have a deficiency or absence of GPIanchored proteins. Central to the pathophysiology of PNH is the absence of two specific GPIanchored proteins, CD55 and CD59 [9,10]. These two proteins are membrane-bound complement inhibitory proteins which function to protect cells from complement attack. Specifically, CD55 regulates the early phase of the complement cascade by inhibiting C3 convertases while CD59 regulates the terminal phase of the complement cascade by inhibiting the formation of
www.expert-reviews.com
10.1586/1744666X.2013.842896
the membrane attack complex [11]. The absence of these complement regulators on the surface of erythrocytes makes these cells susceptible to complement attack leading the clinical signs and symptoms observed in PNH. Complement activation on the surface of the affected cells leads to opsonization with complement fragments and formation of the membrane attack complex. These processes lead to intravascular and extravascular hemolysis [12–14]. Intravascular hemolysis is at the center of the pathophysiology of PNH, while extravascular hemolysis was not recognized until effective therapies to block intravascular hemolysis were developed. Intravascular hemolysis leads to release of free hemoglobin, which in turn scavenges and depletes nitric oxide. It is this depletion of nitric oxide that contributes to the clinical manifestation of PNH including thrombosis, smooth muscle dystonias (e.g., esophageal spasms, male erectile dysfunction) and renal insufficiency [5,6,15]. The severity of symptoms is variable among patients diagnosed with PNH and not all patients require treatment. The only known curative treatment for PNH is bone
Ó 2013 Informa UK Ltd
ISSN 1744-666X
1113
Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
Review
Varela & Brodsky
marrow transplantation; however, this therapy is associated with high morbidity and mortality [16–19]. Until the early 2000’s, the only other available treatment for PNH was management of individual symptoms (i.e., transfusions for anemia, anticoagulation for thrombosis). In 2002, eculizumab, a monoclonal antibody against the complement protein C5, was first used successfully in the treatment of patients with PNH [20]. C5 is the central protein in the complement cascade and its blockade prevents complement attack via the MAC regardless of the pathway of activation. Significant clinical improvements were realized by blocking the terminal pathway of complement with eculizumab in patients with PNH. In the initial pilot study of 11 patients with PNH, treatment with eculizumab reduced intravascular hemolysis, hemoglobinuria, reduced the need for transfusion and showed a positive effect on the quality of life (QoL) in patients with PNH. Subsequently, two Phase III trials (TRIUMPH and SHEPHERD), confirmed the results obtained in the initial pilot study [21,22]. These findings led to the approval of eculizumab for the treatment of patients with PNH by the US FDA and the European Medicines Agency (EMA) in 2007. This article will focus on the pathophysiology of PNH, rationale for targeting the complement system in PNH, the development and clinical use of eculizumab as treatment for PNH and the prospects for further therapies for the treatment of PNH. PIG-A mutation & deficiency of GPI-anchored proteins
The central defect in stem cells and their progeny in patients with PNH is a somatic mutation in the gene encoding for PIG-A. PIG-A is an enzyme involved in the biosynthesis of GPI-anchors, and these anchors are required for the attachment of dozens of proteins to the plasma membrane of cells. The synthesis of GPI anchors is a complex biological process that involves several biochemical reactions and the interaction of more than 20 gene products [23]. PIG-A, in concert with six other enzymes, catalyzes the first step in GPI-anchored biosynthesis. Specifically, it catalyzes the transfer of Nacetylglucosamine (GlcNAc) to phosphatidylinositol (PI) forming GlcNAc-PI. From this crucial step, the synthesis of GPI anchors continues in the endoplasmic reticulum and the completed product is transferred en bloc to the carboxyl terminus of proteins that have a GPI-attachment signal peptide. Early studies of PNH blood cells revealed an absence of specific cell surface proteins. Some of the proteins identified as missing at that time were leukocyte alkaline phosphatase, erythrocyte acetylcholinesterase and later decay accelerating factor (CD55). While absence of the first two proteins did not explain the pathophysiology of PNH, absence of the latter hinted at the mechanism of hemolysis. At the time, it was not known that these proteins were attached to the cell membrane by GPI anchors and it was not until this was determined that the link between PNH and absence of GPI-anchored proteins was established [24,25]. Subsequent studies using cell lines derived from PNH patients determined that the absence of GPI-anchored proteins on the cell surface was due to a defect in the biosynthesis of GPI anchors. Specifically, it was found that cells from PNH patients were unable to carry out the first 1114
step in GPI anchor biosynthesis and were unable to synthesize GlcNAc-PI [26]. Further investigations provided a convincing link between PNH and PIG-A as they revealed that the observed defect in GPI anchor biosynthesis was due to a somatic mutation of the PIG-A gene [27,28]. PNH stem cells & clonal expansion
Deficiency of GPI-anchored proteins occurs in all blood lineages in patients with PNH. The affected cells are clonal and have common PIG-A mutations proving that these mutations arises multipotent hematopoietic stem cells [29,30]. PIG-A mutations have also been described in patients with aplastic anemia (AA), myelodysplastic syndrome (MDS) and in normal healthy controls [30–33]. Similar to PNH, the PIG-A mutations found in patients with AA occur in multipotent hematopoietic stem cells. While AA patients do not show the clinical signs of PNH (hemolysis, smooth muscle dystonias and thrombosis) early in their disease, many AA patients with a PNH clone eventually develop classical PNH. In contrast, MDS patients do not develop PNH. This observation can be explained by the finding that the PIG-A mutations observed in MDS patients and normal controls are transient and have been found to occur in more differentiated colony-forming cells. These differentiated colonyforming cells do not have the self-renewal capacity of multipotent hematopoietic stem cells and are quickly exhausted [30,33]. The mechanisms leading to the clonal expansion and dominance of PNH stem cells remain a topic of continued investigation. The leading hypothesis is that PNH stem cells have a conditional advantage in the setting of an autoimmune attack (e.g., aplastic anemia) [34]. The precise mechanism remains unclear but it may be related, in part, to the absence of GPIanchored ULBP ligands on the surface PNH cells; the receptor for these ULBP ligands in NKG2D, an activating receptor on natural killer cells and CD8+ T cells [35,36]. The role of the complement system in PNH
The complement system is one of the main components of the innate immune system and it is a powerful and effective mechanism that protects the host from invading pathogens. It is made up of more than 35 serum and membrane-bound proteins that interact in a series of tightly regulated, proteolytic steps leading to the induction of a number of biological processes with the main objective of protecting the host from pathogens such as bacteria, viruses, virus-infected cells, parasites and tumor cells (FIGURE 1). In addition, the complement system plays an important role in the disposal of waste (i.e., clearance of immune complexes and apoptotic cells) and it acts as a bridge between innate and adaptive immunity [11,37]. Activation of complement can be achieved via three different pathways: the classical, alternative and lectin. Each of these pathways is initiated by different stimuli, but all lead to the activation of the central complement proteins, C3 and C5. Subsequent steps in the complement cascade result in the activation of a number of complement-dependent mechanisms of high importance for both the innate and adaptive immune responses. Activation of Expert Rev. Clin. Immunol. 9(11), (2013)
Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
PNH & the age of therapeutic complement inhibition
Review
the complement system is tightly reguClassical pathway Alternative pathway lated via soluble and membrane bound and 'tickover' lectin pathway proteins. There are two main points of regulation of the complement cascade, CD55 control of activation of the C3 convertases and control of the formaC3 convertase C3 convertase C3 tion of the membrane attack complex (C3bBb) (C4b2a) (MAC). In erythrocytes, these regulatory functions are carried out by the membrane bound proteins, CD55 and CD59. Amplification loop CD55 inhibits the formation of the C3b C3 convertases as well as accelerates their decay in all pathways of complement activation [38]. CD59 inhibits the terminal pathway and directly prevents the formaC3b Eculizumab C5 tion of the MAC [10]. Both CD55 and CD59 are bound to the cell surface via GPI anchors and are absent in PNH MAC C5a C5b erythrocytes rendering these cell susceptible to complement attack [9,39]. Complement activation on the surface of PNH Extravascular CD59 erythrocytes leads to intravascular and hemolysis extravascular hemolysis by two different mechanisms (FIGURE 1). Central to these mechanisms is the alternative pathway of complement activation. In this pathway, C3 protein spontaneously hydrolyzes and Free hemoglobin leads to the formation of C3 convertase Hemoglobin (this process is also known as ‘tickover’). The mechanism of intravascular hemolysis begins with the increased activity of C3 convertases on the surface of PNH erythrocytes as a result of the lack of CD55. This leads to activation of C3, Intravascular hemolysis C5 and the terminal pathway of complement culminating in the formation of the Figure 1. The complement system, paroxysmal nocturnal hemoglobinuria and MAC. Under normal conditions, formaeculizumab. The complement system is activated via the classical, lectin or alternative tion of the MAC is under the regulation pathways. C3 is activated via C3 convertases. This step is regulated by the action of of CD59. The absence of CD59 on CD55, a GPI-anchored protein. Subsequently, C5 is cleaved into C5a and C5b. C5a mediates a number of biological processes while C5b begins the terminal pathway of complePNH erythrocytes leads to uncontrolled ment and the assembly of the MAC. Deposition of MAC on the surface of cells leads to formation of the MAC resulting in cell lysis. The formation of the MAC is regulated by the action of CD59, another GPIcomplement – mediated intravascular anchored protein. Paroxysmal nocturnal hemoglobinuria (PNH) cells have a deficiency in hemolysis. The mechanism of extravascuGPI-anchored proteins on their cell surface. Absence of CD55 and CD59 leads to unconlar hemolysis begins with increased opsotrolled complement activation on the surface of PNH cells. Deficiency of CD59 increases MAC formation and induces intravascular hemolysis, which is central to the pathophysiolnization of PNH erythrocytes by ogy of PNH. Deficiency of CD55 leads to increased C3 convertase activity and C3dcomplement fragments (mostly C3d). associated extravascular hemolysis. Eculizumab is a humanized monoclonal antibody that This is the result of the lack of CD55. targets C5. By preventing C5 activation, eculizumab prevents the formation of the MAC Opsonized erythrocytes are cleared and leading to a significant reduction in intravascular hemolysis of PNH cells. Use of eculizudestroyed by cells of the reticulomab unmasks a low-level C3-d mediated extravascular hemolysis. MAC: Membrane attack complex. endothelial system [40]. While both CD55 and CD59 are missing on the surface of PNH erythrocytes, it is the absence of CD59 leading to hemolysis, data has demonstrated that unregulated complement unregulated activation of the terminal pathway of complement activation also affects platelets, leukocytes and endothelial cells and the resulting intravascular hemolysis that drives the patho- in patients with PNH. It has been shown that platelets lacking physiology of the disease. In addition to precipitating CD59 produce more procoagulant vesicles [41,42]. Complement www.expert-reviews.com
1115
Review
Varela & Brodsky
activation on macrophages and monocytes, leads to cell injury and release of tissue factor [43]. Most recently, studies have shown evidence of endothelial cell dysfunction in patients with PNH [44,45].
Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
Complement-mediated hemolysis & nitric oxide scavenging
Complement-mediated intravascular hemolysis leads to the release of free hemoglobin into the plasma of PNH patients. Free hemoglobin is normally cleared by haptoglobin, CD163 and hemopexin [12]. The extensive hemolysis seen in PNH overwhelms the clearing mechanisms and leads to accumulation of high levels of free hemoglobin in the plasma. The main consequence of this process is the depletion of nitric oxide (NO). Free hemoglobin is a potent NO scavenger as the two undergo a fast and irreversible reaction that results in the production of nitrate (NO3) and methemoglobin. Normally, NO is synthesized by endothelial cells and functions to maintain smooth muscle relaxation and inhibit platelet activation and aggregation. Additionally, erythrocyte arginase is released as a result of hemolysis leading to depletion of L-arginase, a precursor of NO, resulting in decreased production of NO. The absence of NO as a result of scavenging by free hemoglobin and decreased production due to low L-arginase levels leads to deregulation of muscle tone, endothelial dysfunction and inappropriate activation of platelets (FIGURE 2). In turn, these processes lead to smooth muscle dystonias and propensity for thrombosis. Diagnosis & clinical aspects of PNH
PNH is diagnosed based on clinical symptoms and laboratory studies. Clinical signs can include symptoms from hemolytic anemia, thromboses, hemoglobinuria, smooth muscle dystonias (e.g., esophageal spasms, male erectile dysfunction), abdominal pain, arterial/pulmonary hypertension and renal insufficiency [1,4,5,15]. Laboratory studies used for the diagnosis of PNH include basic hematological tests including complete blood counts, reticulocyte count, lactate dehydrogenase (LDH), haptoglobin levels, flow cytometric evaluation of GPI anchored proteins [46–48] and sometimes bone marrow biopsy. PNH can be divided into three subcategories, classical PNH, PNH in the setting of another specified bone marrow disorder and subclinical PNH. Patients with classical PNH have signs and symptoms of intravascular hemolysis (i.e., increased reticulocyte count, increased serum LDH) and have no evidence of another bone marrow disorder. While not required for the diagnosis of PNH, bone marrow evaluation in these patients reveals a hypercellular bone marrow with erythroid hyperplasia but without karyotypic abnormalities [2,49]. Patients with PNH in the setting of another specified bone marrow disorder have clinical and laboratory evidence of hemolysis and also have another defined bone marrow abnormality (presently or historically). Patients with subclinical PNH have no evidence of hemolysis but have a small clone of PNH cells. These patients can progress to clinically significant PNH, as it is the case with patients 1116
with acquired AA. At diagnosis, most patient with AA have 60% also appear to be at higher risk for thrombosis [5,15]. The mechanisms leading to thrombosis in PNH are not fully elucidated, but several processes have been associated with the prothrombotic state. These processes include NO scavenging by free hemoglobin leading to deregulation of platelet activation and aggregation, production of C5a during complement activation leading to upregulation of tissue factor, generation of PNH plateletderived microvesicles with phosphatidylserine, a known in vitro procoagulant [51,42], release of tissue factor from macrophages and monocytes damaged by complement [43] and endothelial cell dysfunction leading to an increased risk of thrombosis [45]. Treatment of PNH before the development of eculizumab
Treatment of PNH is reserved for those with significant disease manifestations such as disabling fatigue, thromboses, transfusion dependence, or renal insufficiency [2]. Bone marrow transplantation is the only known curative treatment for PNH. Different studies have reported a 2-year survival of 56% [52] and 10-year survival of 57% [18]. More recently, nonmyeloablative conditioning regimens have been used with both HLA-identical [53] and HLA-haploidentical donors [19]. Currently, bone marrow transplantation is only recommended for patients with severe manifestations (life-threatening cytopenias) who do not respond to drug therapy and in countries where eculizumab therapy is not available [54]. Most recently a study of 211 patients recommended against the use of allogeneic stem cell transplantation in patients with life-threatening thromboembolism given the availability of eculizumab [55]. Prior to the development of eculizumab and besides bone marrow transplantation, the treatment of PNH was focused on supportive therapy. For the ongoing hemolytic anemia, patients received red blood cell transfusions, iron and folic acid supplementation [56]. Corticosteroids were successful at decreasing complement-mediated hemolysis in some patients, but their toxicity and lack of reliable effect limit their use. The use of prophylactic anticoagulation in patients with PNH remains a topic of controversy. While some groups have suggested Expert Rev. Clin. Immunol. 9(11), (2013)
PNH & the age of therapeutic complement inhibition
Review
Free hemoglobin
Fe2+
Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
NO
O2
NO NO Complement-mediated intravascular hemolysis
O2 O2
NO
O2
Nitric oxide scavenging
+
NO3-
Fe3+ Nitric oxide depletion
Impaired regulation of smooth muscle tone
Platelet activation and aggregation
Smooth muscle dystonias (abdominal pain, esophageal reflux, erectile dysfunction
Thrombosis
Figure 2. Complement-mediated intravascular hemolysis leads to nitric oxide scavenging and depletion in paroxysmal nocturnal hemoglobinuria. Complement-mediated intravascular lysis of paroxysmal nocturnal hemoglobinuria erythrocytes leads to leakage of free hemoglobin into the plasma. In turn, free hemoglobin scavenges and depletes NO. Depletion of NO leads to dysregulation of smooth muscle tone leading to the dystonias associated with paroxysmal nocturnal hemoglobinuria (i.e., esophageal spasms, erectile dysfunction). In addition, depletion of NO leads to platelet activation and aggregation, this is believed to contribute to the increased risk of thrombosis. NO: Nitric oxide.
primary prophylaxis with warfarin in PNH patients with a granulocyte clone of more than 50% and no contraindications to anticoagulation [57], other groups suggests that guidelines for primary prophylaxis cannot be made based on the available data [49]. Most recently, a study from the South Korean registry, concluded that patients with LDH levels >1.5 ULN at diagnosis were at a significantly higher risk of thromboembolism and that the combination of increased LDH (>1.5 ULN) with the clinical symptoms of abdominal pain, chest pain, dyspnea or hemoglobinuria was associated with a greater increased risk for thromboembolism [58]. www.expert-reviews.com
Eculizumab: the first anti-complement drug
The complement system has been implicated in the pathogenesis of a number of human diseases. Searching for therapeutic agents targeting the complement system has been a subject of much discussion and research. An ideal anti-complement therapeutic would target the complement cascade at the point of convergence of all of the activating pathways and before the cleavage of C5. Therefore, developing therapeutics targeting the C5 protein would be the most rational approach. In the 1990’s, a number of murine anti-human C5 monoclonal antibodies were screened with the goal of finding an antibody that 1117
Review
Varela & Brodsky
Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
Lactate dehydrogenase (U/l)
3000 2500 2000 Eculizumab (n = 43) 1500
Placebo (n = 44)
have an important role in treatment of the disease. It was expected that blockade of C5 with eculizumab would lead to inhibition of the terminal pathway of complement, prevention of MAC assembly and inhibition of complementmediated intravascular hemolysis (FIGURE 1)
1000
Eculizumab for PNH 500
Initial Phase II pilot study
The first study of eculizumab in patients with PNH was started in May of 0 10 20 30 40 50 2002 [20]. It was a 3-month, open-label, Time, weeks Phase II pilot study which enrolled eleven PNH patients (six men and five women), Figure 3. Treatment with eculizumab leads to a significant reduction in complement-mediated intravascular hemolysis. Mean levels of lactate dehydrogenwho had a detectable GPI-deficient hemaase (LDH) reflect the degree of hemolysis. The dashed line represents the upper limit of topoietic clone and had received at least the normal range for LDH (103–230 U/l). In patients treated with eculizumab, the level of four red cell transfusions in the preceding LDH was rapidly and significantly reduced to levels just above the normal range. In the 12 months. The study investigated the placebo group, the mean LDH level remained highly elevated. The arrow represents the effect of eculizumab on complementtime of transition of placebo-treated patients to treatment of eculizumab under the extension study following the initial TRIUMPH. Initiation of eculizumab treatment in mediated hemolysis as determined by placebo-treated patients led to a significant decreased in LDH to near normal levels. LDH, haptoglobin, bilirubin, hemoglobin, Reproduced with permission from [59]. reticulocyte counts, hemoglobinuria and need for transfusions. In addition, it measwould both inhibit the generation of C5a and the formation of ured the effect on the proportion of GPI-deficient cells as well the MAC. One antibody, m5G1.1, was selected out of more as any effect on the QoL of the study participants. Patients 30,000 screened [59]. Subsequently, this antibody was modified for were given 600 mg of eculizumab iv. weekly for 4 weeks, foluse as a human therapeutic. First, in order to minimize immuno- lowed by a 900 mg dose on week 5 and then by a 900 mg genicity, m5G1.1 was humanized by grafting its complementarity- dose every other week through week 12. In this pilot study, determining regions in human heavy and light chain antibody eculizumab was shown to significantly decrease complementframeworks. In addition, in order to eliminate antibody-mediated mediated hemolysis as evidenced by a significant decreased in effector functions (i.e., activation of complement, initiation of Fc LDH levels (FIGURE 3), transfusion requirements, hemoglobinureceptor-mediated pathways), the heavy chain constant region of ria and number of paroxysms. Interestingly, the LDH levels m5G1.1 was replaced with components of both human IgG2 and decreased to mean above the highest limit of normal, indicatIgG4 constant regions [59]. These IgG subtypes were chosen ing some persistent low-level hemolysis. The QoL of particibecause IgG2 is known to not bind Fc receptors and IgG4 does pants was also evaluated and found to be significantly not activate complement. The resulting antibody was the first improved. Importantly, this pilot study also investigated the anti-complement drug, eculizumab, a humanized monoclonal safety, pharmacokinetics and immunogenicity of eculizumab. antibody targeting the complement protein C5, which inhibited There were no treatment-related deaths or thrombotic events. the generation of C5a, prevented the formation of the MAC, was One of the main concerns at the time was the increased risk not immunogenic and did not activate complement or Fc for infection, especially by encapsulated organisms. All study receptor-mediated pathways. participants were vaccinated against Neisseria meningitidis Preclinical in vivo studies of eculizumab were carried out in prior to receiving eculizumab and no infectious complications mouse models of arthritis and lupus. Of note, given the were reported. The study also showed that therapeutic levels humanized nature of eculizumab, a mouse anti-C5 antibody of eculizumab were reached and maintained throughout the that demonstrated similar activity to eculizumab was used study period. No antibodies against eculizumab were detected instead. In both models, blockade of the complement cascade in any of the participants. Following the initial pilot study, at C5 led to amelioration of disease and prolonged survival of all participants continued treatment in a 52-week extension mice [60]. The first human Phase I trials of eculizumab were study [61]. The yearlong study confirmed all the findings of performed in patients with rheumatoid arthritis, membranous the pilot study. nephritis and systemic lupus erythematosus (SLE). All these studies were Phase I trials aimed at determining the safety and Phase III trials dosing of eculizumab in humans rather than testing its efficacy After the compelling data from the initial pilot study, eculizuagainst disease [59]. While no Phase I trials were carried out in mab received orphan designation for PNH and two Phase III PNH patients, it quickly became apparent that this drug could trials were started in the USA, Europe and Australia. 0
1118
Expert Rev. Clin. Immunol. 9(11), (2013)
Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
PNH & the age of therapeutic complement inhibition
The first study was TRIUMPH; a double-blind, randomized, placebo-controlled, multicenter, Phase III trial [21]. Participants were required to have a PNH type III erythrocyte proportion of at least 10%, LDH levels of at least 1.5-times the upper limit of normal, a platelet count of at least 100,000 per mm3, and to have received at least four transfusions during the previous 12 months. A total of 87 patients (35 men and 52 women) were enrolled in the trial. Patients were randomized into an eculizumab group (n = 43) or a placebo group (n = 44). Patients in the treatment group received 600 mg infusion of eculizumab weekly for 4 weeks, followed by a 900 mg infusion on week 5 and then a 900 mg infusion every other week until week 26. The study examined the effect of eculizumab on hemoglobin levels, transfusion requirements, complement-mediated hemolysis and participant’s QoL. Similar to the pilot the study, TRIUMPH demonstrated the efficacy of eculizumab in the treatment for PNH. It demonstrated a stabilization of hemoglobin levels in 49% of patients in the treatment group compared to 0% in the placebo group (p < 0.001). Transfusion requirements were significantly different in the treatment versus placebo groups. Within 6 months prior to the initiation of the study treatment group patients received 413 units of pack red blood cells versus 417 units for the placebo group. In marked contrast, by week 26 of the study, the treatment group had received 131 units versus 482 units for the placebo group. Hemolysis was significantly decreased in treated patients as evidenced by decreased LDH levels and increased proportions of PNH type III erythrocytes. QoL was also significantly improved in the treatment group compared to the placebo group. There were no treatment related deaths and no serious adverse events were attributed to the treatment. The most common adverse events reported in the eculizumab group were headache and back pain. One patient in each group had low, detectable levels of antibodies against eculizumab, but they were not clinically significant. All study participants were vaccinated against Neisseria meningitides prior to receiving eculizumab and no infectious complications were reported. The second study was SHEPHERD, an open-label, Phase III trial designed to investigate the long-term safety and efficacy of eculizumab by relaxing the inclusion criteria of the TRIUMPH trial [22]. In contrast to the eligibility criteria for TRIUMPH, this trial allowed patients with platelet counts as low as 30,000 per mm 3 and patients were only required to have had at least 1 transfusion in the previous 2 years. Similar to TRIUMPH, participants were required to have a PNH type III erythrocyte proportion of at least 10% and LDH levels of at least 1.5-times the upper limit of normal. A total of 97 patients (48 men and 49 women) were treated in this study. Patient received intravenous infusion of 600 mg of eculizumab weekly for 4 weeks followed by a 900 mg dose a week later and then 900 mg every 14 days for a total of 52 weeks. The study measured adverse events (AEs) possibly related to eculizumab, its effect on hemolysis, fatigue and quality of life of enrolled patients. There was one death www.expert-reviews.com
Review
in the study due to thrombosis, but it was considered unrelated to the study drug. The most common AEs reported were headaches, nasopharyngitis and upper respiratory tract infection which occurred in 52, 32 and 30% of patients, respectively. There were seven serious AEs considered to be possibly related to eculizumab, fever (2), headache (1), abdominal distension (1), viral infection (1), anxiety (1) and renal impairment (1). None of these, however, were considered probably or definitely related to eculizumab. As in previous trials, all patients were immunized against Neisseria meningitides prior to receiving eculizumab. Eighty-nine patients reported at least one infection during the study period. The majority of infections were considered mild to moderate and not related to the study drug. The study also showed that treatment with eculizumab led to a rapid, sustained reduction in hemolysis as determined by significantly decreased LDH levels and increased proportions of PNH type III erythrocytes. The participants also reported a significant decrease in fatigue and improvement in quality of life. Interestingly, 71% of patients in the SHEPHERD study would not have met the inclusion criteria for TRIUMPH; the SHEPHERD trial demonstrated that eculizumab treatment was applicable to broad population of patients with PNH. Additional studies
The main findings from the pilot study, TRIUMPH and SHEPHERD were that eculizumab was an effective and safe treatment for PNH. It reduced the complement-mediated hemolysis, minimized transfusion requirements and improved the QoL of treated patients with reproducible effectiveness and few adverse events. Following the completion of the pilot, TRIUMPH and SHEPHERD studies a total of 187 patients were enrolled in an extension Phase IIIb trial to assess the long-term efficacy safety and efficacy of continued eculizumab treatment [62]. All three original trials followed the same dosing schedule and the patients enrolled in the extension trial continued to receive the maintenance dose of 900 mg of eculizumab every 2 weeks for a total of 66 months. This long-term study was a post hoc analysis and reported that there was a significant reduction in hemolysis, reduction in transfusion dependence, decrease in thromboembolic events and improvement in renal function. Patients treated long-term with eculizumab had a 3-year survival of 97.6% [63]. The AEs were unchanged from the previously reported trials and most commonly included headache, nasopharyngitis and upper respiratory tract infections. Fortypatients reported infectious complications during treatment. Of those, two patients developed meningococcal sepsis and both were treated without complications. Other studies evaluated the effect of eculizumab on other aspects of the pathophysiology of PNH using the same cohort of patients as the three original trials. One study investigated the effect of long-term treatment with eculizumab on thromboembolism (TE) [64]. This study included patients from the 1119
Expert Review of Clinical Immunology Downloaded from informahealthcare.com by Michigan University on 10/29/14 For personal use only.
Review
Varela & Brodsky
original pilot trial, TRIUMPH, SHEPHERD and the Phase IIIb extension study. Thromboembolism rate with eculizumab was compared with pretreatment rate in the same patients. The study found a relative reduction of 85% in TE event rate during eculizumab treatment (TE rate before eculizumab = 7.37 events/100 patient years vs, TE rate with eculizumab = 1.07 events/100 patient years; p < 0.001). In patients treated with anticoagulants, there was a relative reduction of 94% with eculizumab (TE event rate was reduced from 10.61–0.62 events/100 patient years, p < 0.001). Another study investigated the effect of eculizumab on NO depletion, dyspnea and measures of pulmonary hypertension [65]. This study was carried out with patients from the TRIUMPH trial only. Treatment with eculizumab significantly reduced NO depletion, dyspnea and decreased the proportion of patients with elevated pro-brain naturetic peptide (proBNP, a marker for pulmonary vascular resistance and right ventricular dysfunction in hemolytic diseases). Treatment with eculizumab beyond the initial trials
Shortly after the completion of TRIUMPH and SHEPHERD, eculizumab became the first drug approved by the FDA and EMA for the treatment of complement-mediated hemolysis in patients with PNH. Most of the published literature regarding the use of eculizumab in patients with PNH has been derived from studies of 187 patients that were enrolled in the clinical trials that lead to FDA approval. To date, few studies have evaluated eculizumab outside the context of a clinical trial. One study reported on 79 patients treated with eculizumab at a single center in the UK [66]. Of the 79 patients, 34 were enrolled in one of the original clinical trials. The patients received eculizumab per the dosing and treatment schedule used in the previous trials. The mean follow-up was 39 months. The study compared survival of PNH patients treated with eculizumab to that of patients with PNH in the 7 years prior to the use of eculizumab and that of the general population. The 5-year survival of patients treated with eculizumab was significantly better than that of patients not treated with eculizumab (95.5 vs 66.8%, respectively, p = 0.0062) and comparable to the survival of the general population (p = 0.46). A second study was a retrospective, single center study that evaluated the response of 30 patients with PNH to treatment with eculizumab in the USA [67]. Of note, out of the 30 patients, five were enrolled on the TRIUMPH or SHEPHERD trial. The patients received eculizumab using the same dosing regimen as the published trials. The patients were categorized based on response criteria. A complete response (CR) was defined as transfusion independence with normal hemoglobin for age and sex, absence of symptoms (i.e., thromboses, smooth muscle dystonias) and a LDH
![[Paroxysmal nocturnal hemoglobinuria].](/assets/img/hold.png)
