The Management of the Haemophilic Arthropathy

Pathogenesis of the Haemophilic Arthropathy

Daniela Melchiorre¹, *, Silvia Linari², Fabrizio Matassi³, Giancarlo Castaman²

¹ Rheumatology Unit, University of Florence, Florence, Italy

² Center for Bleeding Disorders, Azienda Ospedaliero-Universitaria Careggi, Florence, Italy

³ Orthopaedic Clinic, University of Florence, Florence, Italy

Abstract

Joint damage due to recurrent bleedings in Haemophilia is the cause for long-term disabilities. The pathogenetic mechanism of haemophilic arthropathy is multifactorial and includes inflammatory synovium-mediated and degenerative cartilage-mediated phenomenons, in addition to neoangiogenesis and bone loss. Free blood in the joint has a direct effect on cartilage and synovium, and the deposit of iron appears to play a pivotal role. Iron may promote the apoptosis of chondrocytes by catalyzing the formation of oxygen metabolites. Iron may also act on the synovial membrane by favouring its proliferation through the induction of proto-oncogenes involved in cellular proliferation and stimulation of inflammatory cytokines. Such degenerative and inflammatory processes occur concomitantly, but also independently. A reduction of bone mineralization is usually present as a part of the articular damage associated to a multifactorial mechanism: it seems that the molecular triad (osteoprotegerin/Receptor activator of nuclear factor kB/Receptor activator of nuclear factor kB ligand) probably plays a major role, inducing osteoclastic differentiation and maturation. These processes finally result in a fibrotic and irreversible altered joint, feature of haemophilic arthropathy.

Keywords: Arthropathy, Haemophilia, Haemarthrosis, Haemosiderin, Neoangio-genesis, Osteoporosis, Synovitis.

* Corresponding author Daniela Melchiorre: Rheumatology Unit, University of Florence, Florence, Italy; Tel: +39 0552751688; Email: daniela.melchiorre@unifi.it

INTRODUCTION

Haemophilia A and B are rare X-linked recessive bleeding disorders characterized by the absence or functional defect of clotting factor VIII (FVIII) or factor IX (FIX) respectively. The hallmark of such disease is represented by musculoskeletal bleedings, particularly haemarthrosis, leading to orthopaedic complications. Joint bleeding is the most common and potentially most disabling manifestation of severe Haemophilia (i.e. plasma FVIII or FIX <1U/dL) [1]. In

nearly half of all children affected by severe Haemophilia, the initial haemartrosis occurs during the first year of life [2], and 90% of patients experience at least a joint bleeding before the age of 4.5 years [3]. Eighty per cent of joint bleedings involve knees, elbows, and ankles [4], and patients often develop multiple target joints. Although blood is rapidly cleared from the joint space also by the replacement with the missing factor, the pathologic process still continues, resulting in both clinical and radiographic changes. Recurrent bleedings cause an irreversible joint damage with progressive functional impairment [5], chronic pain [6], and heavy impact on quality of life [7]. Haemarthrosis can be prevented or controlled by the prophylactic administration of clotting factor concentrates. Compared with an on-demand treatment strategy, a primary prophylactic treatment (i.e. the regular continuous treatment initiated in the absence of documented osteochondral joint disease and started before the second clinically evident large joint bleed in children >3 years) leads to better musculoskeletal outcomes, as clearly established [8-11]. [8]. However despite such strategy, joint bleedings and related damages may recur and the haemophilic arthropathy (HA) may realize, as confirmed by the radiographic evidence by the age of 6 in some subjects who had no bleeding or few subclinical haemarthroses [11].

The mechanism of the progressive joint damage in patients with Haemophilia is still relatively unclear, but recurrence and persistence of blood into the joint cavity is the key factor responsible for synovial and cartilage changes [12, 13]. Increasing evidences of a close relationship between the type of mutation (null and/or missense mutations), bleeding, inflammatory process, and neoangiogenesis are emerging and suggesting that iron, cytokines, and neoangiogenic factors can initiate synovial and early cartilage damages with molecular changes and perpetuation of a chronic inflammatory condition [14].

From Bleeding to Synovial and Cartilage Damage

Bleeding into a joint exposes synovial cells to blood and its components including iron that plays a pivotal role in joint damage [15] (Fig. 1). The progressive accumulation over time of iron as haemosiderin (normally removed from the by synovial macrophages) represents the trigger for synovial inflammation [16]. Haemosiderin deposits are crucial in the early stages of HA, triggering synoviocyte hypertrophy (resulting in villi), neoangiogenesis, and release of hydrolytic enzymes from synovial cells. Iron up-regulates the expression of proinflammatory cytokines, as interleukin-6 (IL-6), IL-1alpha, IL-1beta and tumor necrosis factor-alpha (TNF-alpha) in synovial cells and induces the regulator genes c-myc and MDM2 expression, resulting in synovial proliferation [17, 18]. Another effect of haemosiderin is the lymphocytes infiltration of the synovial membrane with subsequent inflammatory changes. Moreover, different proinflammatory cytokines released by synovial cells may inhibit the formation of human cartilage matrix [15]. Synovitis is one of the earliest macroscopic effect of a target joint and it is not always easily distinguishable from a clinical point of view from haemarthrosis. Synovitis is an inflammatory process involving synovial tissue, characterized by hypertrophy, migration of inflammatory cells, and a high degree of neoangiogenesis [18-22].

Fig. (1))

Mechanisms of blood-induced joint damage in Haemophilia: the role of iron (Fe²+) interacting with Hydrogen peroxide (H202), macrophages’activation (Mo/Mö), matrix metalloproteinases (MMPs), and involvement of several cytokines.

Recurrent bleedings and synovitis rapidly evolving into joint damage can be considered two different aspects of HA. Intense, chronic effusion of the affected joint after one or several haemarthrosis typically occurs in the early stages of HA [23, 24]. Synovitis can lead to further bleedings with transformation of an acute process in a chronic disease.

Also the presence of free blood in a target joint has a direct harmful effect on cartilage, resulting in adverse changes in chondrocyte activity [25]. Moreover, these alterations may occur before the synovial inflammation becomes evident. Human articular cartilage consists of a relatively small number of chondrocytes embedded in a relatively large amount of extracellular matrix that consists mainly of collagen and proteoglycans. There is a continuous turnover of these components, with a delicate balance between synthesis and breakdown [26].

A pivotal role in the whole process is played by the high-weight molecular complex called inflammasome. The inflammasome controls the maturation and the secretion of IL-1beta by means of the activation of caspase 1. The inflammasome constituents are the pattern-recognition receptors (PRRs), including the Toll-like receptors (TLRs) and lectins (CTLs), which analyze the extracellular environment and are associated with pathogen-associated molecular patterns (PAMPs). PRRs also include the intracellular NOD-like receptors (NLR) which recognize both PAMPs that harmful signals caused by danger associated molecular patterns (DAMPs) as demonstrate by Mendonça et colleagues [27].

In vitro studies have shown that a relatively short exposure (4 days, the expected natural evacuation time of blood from a human joint) of human cartilage to whole blood in concentrations up to 50% (blood concentration during haemarthroses are expected to approach 100%) induces long-lasting damaging effects [28]. The marked inhibition of matrix formation (proteoglycan synthesis) and increased breakdown, i.e. release of matrix components (proteoglycan release) result in a progressive loss of matrix, caused by the induction of apoptosis of chondrocytes by hydroxyl radicals formed upon exposure to blood [29]. Hydroxyl radicals are formed when hydrogen peroxide production by chondrocytes is increased upon stimulation by pro-inflammatory cytokines, such as IL-1beta, originating from activated blood monocytes/macrophages present in the blood within the joint. Hydrogen peroxide reacts with haemoglobin-derived iron from damaged and phagocytosed red blood cells close to chondrocytes. This triggers the formation of radicals that induce apoptosis and consequently an irreversible inhibition of cartilage matrix synthesis [25, 29, 30]. Canine in vivo studies have corroborated these findings [31] and demonstrated that immature cartilage is more susceptible to blood-induced damage than mature cartilage [32]. Joint bleeding leads to initially independent adverse changes in synovial tissue, articular cartilage, and consequently subchondral bone. Taken together, the mechanism of blood-induced joint damage includes both degenerative (cartilage-mediated) and inflammatory (synovium-mediated) components. Although influencing each other, these processes also occur independently [33].

Neoangiogenesis

Neoangiogenesis associated with the recruitment of bone marrow-derived progenitors is a critical, independent mechanism involved in the development and maintenance of HA. Neoangiogenesis is also implicated in tumor growth and inflammatory arthritis [14, 34, 35]. The proangiogenic vascular endothelial growth factor (VEGF) is the principal signaling molecule in angiogenesis and can be induced by hypoxia and different cytokines through interaction with its receptors, VEGFR1 and 2. Similarly to other joint diseases, sharing histological similarities with HA, the synovial pannus has enhanced oxygen demand with evidence of de novo blood vessel formation of the synovium. A four-fold elevation in different proangiogenic factors as VEGF, stromal cell-derived factor-1 (SDF-1), and metalloproteinase 9 (MMP-9) has been recorded. Also pro-angiogenic macrophage/monocyte cells (VEGF+/CD68+ and VEGFR1+/CD11b+ and VEGF/CD68+) in synovium and peripheral blood of haemophilic subjects were observed. A significant increase of VEGFR2/AC133+ endothelial progenitor cells and CD34/ VEGFR1+ haemopoietic progenitors cells was also demonstrated [34, 36]. Human synovial cells, when incubated with haemophilic sera, up-regulated hypoxia-inducible factor 1alpha (HIF1A) mRNA, implicating hypoxia in the neoangiogenesis process [36]. Moreover, an increased microvessels density has been shown by immunofluorescence in synovial cells from patients with end-stage HA [34]. This suggests that also late stages of HA can be characterized by an active neoangiogenesis.

From Bleeding to Osteoporosis

Osteoporosis is a disorder characterized by decreased bone mass and microarchitectural deterioration, resulting in loss of bone strength and fragility fractures [37]. Such condition has been recently recognized as a severe comorbidity in Haemophilia [38, 39].

The pathogenesis of low bone mineral density in subjects with Haemophilia is multifactorial. Primary key factors include: prolonged immobilization [38-40]; lack of weight-bearing exercises and failure to achieve an optimal strength during growth [41]; excessive bone resorption resulting in loss of bone mass and failure to replace the lost bone due to defects in bone formation [42, 43]. Other factors such as Human Immunodeficiency Virus (HIV) or hepatitis C virus (HCV) infections, and their treatments, may be independently associated with decreased bone mineral density [38, 42, 44]. The mechanism and pathways by which blood in the joint cavity causes bone quality and quantity depletion have not been to date fully elucidated [45, 46]. However, members of the TNF receptor superfamily probably play a major role. Effectively, the molecular triad osteoprotegerin/Receptor activator of nuclear factor kB/Receptor activator of nuclear factor kB ligand (OPG/RANK/RANKL) that tightly controls the bone turnover is a crucial parameter of bone biology [47, 48]. RANKL is a transmembrane ligand mainly expressed on osteoblasts/stromal cells in the bone microenvironment. RANKL exists either as a cell-bound form or a truncated ectodomain variant derived by enzymatic cleavage of the cellular form (soluble RANKL, sRANKL). It binds to its receptor RANK expressed at cell surface of osteoclast precursor, possibly of the macrophage lineage, and induces osteoclastic differentiation and maturation, leading to bone resorption. In synovial membrane, RANKL is expressed by fibroblast-like synoviocytes (synoviocytes type B), and by activated T cells and may induce osteoclastogenesis, through a mechanism enhanced by several cytokines (TNF-alpha, IL1, and IL17) that promote both inflammation and bone resorption [49, 50]. OPG, also a member of the TNF receptor family, acts as a decoy receptor for RANKL, and competes for binding of RANKL to RANK [51-53]. By this mechanism, OPG negatively regulates osteoclast differentiation, activity, and survival both in vivo and in vitro [48, 50, 53, 54]. RANKL inhibits osteoclast apoptosis whereby OPG acts as antagonist [49]. OPG is predominantly found in macrophages of the intimal synovial lining and in endothelial cells, where is complexed with von Willebrand factor within the Weibel-Palade bodies [54]. Variations in the balance between OPG and RANKL leads to pathological bone changes. Osteoclasts precursors (OCPs) are derived from haemopoietic (monocyte) progenitors in the spleen and liver migrating from blood into bone where they fuse with one another to form multinucleated osteoclasts. Blood neutrophils in the joint create an inflammatory environment that produces IL-1, IL-6, RANKL, and TNF-alpha [55, 56]. TNF increases the proliferation and differentiation of OPCs [57]. TNF also inhibits the production by bone marrow stromal cells of stromal cell-derived factor 1, which, in turn, increases the release of OPCs from the bone marrow [58]. RANKL synthesized by reactive lymphocytes in the joint binds to RANK stimulating osteoclasts to resorb bone (Fig. 2).

Fig. (2))

The role of the molecular triad osteoprotegerin/RANK/RANKL in bone remodelling.

As key regulators of bone remodeling, serum levels of OPG/RANKL were analysed in a wide population of patients with haemophilia, together with the expression of the triad OPG/RANK/RANKL in synovial tissue of adult patients with Haemophilia, undergoing to knee replacement surgery [59]. OPG levels in all Haemophilia A patients were decreased and a strong expression of RANK and RANKL was found. A strict correlation between instrumental findings and severity of HA, according to the World Federation of Haemophilia orthopaedic joint scale (WFH score) [60], Petterson [61], and ultrasound score [62] was also observed. The biochemical markers of bone turnover in the synovial tissue of haemophiliacs indicate an osteoclastic activation, not counteracted by OPG. In fact, RANK and RANKL were found to be strongly expressed in the synovium. Instead, the expression of OPG was dramatically reduced in synovial tissue. The absence of OPG in synovial tissue suggests that the balance is shifted versus osteoclastic activity. In conclusion, osteoclastogenesis seems to be activated in synovial tissue of haemophilic patients, and it may be mediated by an inflammatory milieu (Fig. 3).

Fig. (3))

Expression of receptor activator of nuclear factor-κB (RANK), RANK ligand (RANKL), and osteoprotegerin (OPG) in synovial tissue from patients with Haemophilia A and osteoarthritis. Representative microphotographs of tissue sections subjected to immunoperoxidase staining for RANK, RANKL, and OPG (brownish-red color) and counterstained with hematoxylin are shown.

Arthropathy in Haemophilia A and B: Some Differences

Haemophilia A and B are considered clinically indistinguishable, sharing recurrent joint bleeds as hallmark of a severe disease. However some evidences suggest that Haemophilia B may be less severe than Haemophilia A [1, 63-65]. In a recent study, a large population of patients with Haemophilia A and B was evaluated by using clinical, imaging, and biochemical markers [66]. WFH score and US score were significantly worse in patients with Haemophilia A than the others, matched for age, even with similar frequency of haemarthroses. Notwithstanding the equivalent degree of clotting deficiency, the number of haemarthroses was significantly less in Haemophilia B patients. The lower serum OPG and sRANKL levels in Haemophilia A are in keeping with more severe forms of arthropathy and clinical outcomes. Similar significance can be attributed to the reduced expression of OPG with the marked expression in the synovial tissue of RANK and RANKL in Haemophilia A. Moreover, the histological analysis in synovial tissue of patients affected by Haemophilia B underlines the differences of the expression of RANK/RANKL/OPG triad with respect to Haemophilia A. Effectively, the increased expression of OPG and the lower number of patients undergoing arthroplasty confirm that the arthropathy may be less severe.

CONCLUDING REMARKS

Joint bleeding is the most common and potentially most important long-term disabling manifestation of severe Haemophilia. Recurrent joint bleedings cause irreversible articular damages with progressive functional impairment. The subsequent release of iron from destroyed red cells has a direct pathogenetic effect on cartilage, synovium, and bone. Different cytokines play a crucial role in blood-induced arthropathy inducing an overreaction and leading to irreversible damages independent from the bleeding. These processes finally result in an altered, restructured, and not functional joint, with risk of ankylosis, feature of the classic haemophilic arthropathy that tends to be more severe in Haemophilia A than in Haemophilia B.

CONFLICT OF INTEREST

The authors confirm that they have no conflict of interest to declare for this publication.

ACKNOWLEDGEMENTS

Declared none.

REFERENCES

Pharmacokinetic Approach to the Treatment of Haemophilia

Giancarlo Castaman*, Maria Chiara Susini

Center for Bleeding Disorders, Department of Oncology, Azienda Ospedaliero-Universitaria Careggi, Florence, Italy

Abstract

Pharmacokinetic (PK) has improved our knowledge about the most appropriate dosing and timing of administration of FVIII/FIX concentrates in patients with Haemophilia. However, although several studies have recently addressed the relevance of PK of clotting factors, usual practice is still mostly based on empiric approaches since individual PK estimation is difficult to obtain unless the patient is formally enrolled in a study. In fact, several plasma samples collected over several hours and/or days are required to establish a half-life curve confidently and this may be a relevant problem, especially in children. Recently however population PKs has emerged as an important tool to overcome this drawback. Targeted prophylaxis could take advantage of knowing the individual response to factor concentrate administration. On the clinical ground, age and body weight (BW) are roughly used to guide dosing because usually in vivo recovery is lower and clearance is faster in children than in adults.

Keywords: Factor VIII, Factor IX, Haemophilia A, Haemophilia B, Pharmaco-kinetics.

* Corresponding author Giancarlo Castaman: Center for Bleeding Disorders, Department of Oncology, Azienda Ospedaliero-Universitaria Careggi, Florence, Italy; Tel: +39557947587; E-mail: castaman@aou-careggi.toscana.it

INTRODUCTION

Replacement treatment with clotting factor concentrates (factor VIII – FVIII or factor IX – FIX) has dramatically improved Haemophilia care and prognosis [1]. In conjunction with the evolution of products and therapeutic regimens, the important role of pharmacokinetics (PK) has been also increasingly recognized. Methods for PK evaluation have been developed, progressively becoming more and more accurate. From the 1970s, it has been clearly shown that three times per week treatment was much better than once-weekly for prevention of bleeding [2, 3]. Subsequent careful PK studies showed the benefits of PK plotting and implementation in Haemophilia prophylaxis providing hints to a personalized

prophylaxis in an attempt to strengthen efficacy without increasing the costs [4, 5]. Substitutive treatment for haemophilia is expensive, but inadequate treatment worsens quality of life by increasing morbidity and late sequelae and eventually increasing the associated costs. As in most fields of medical treatment, variability in response among patients is critical and thus tools to optimize clotting factor usage should be always pursued. The required dose should be administered according to the clinical setting (treatment of acute bleeding, surgical prophylaxis or regular prophylaxis), the degree of the factor deficiency, the site and severity of bleeding. On this basis the application of pharmacokinetic analysis would provide the clinicians with more accurate information to tailor patient treatment [6-9].

Pharmacokinetics: Basic Principles

The dose–response relationship for a