Blood Reviews 29 S1 (2015) S9 S18. Contents lists available at ScienceDirect. Blood Reviews. journal homepage:

Transcription

1 Blood Reviews 29 S1 (2015) S9 S18 Contents lists available at ScienceDirect Blood Reviews journal homepage: Recombinant activated factor VII in the treatment of bleeds and for the prevention of surgery-related bleeding in congenital haemophilia with inhibitors Elena Santagostino a, *, Miguel Escobar b, Margareth Ozelo c, Luigi Solimeno d, Per Arkhammar e, Hye Youn Lee f, Gabriela Rosu f, Paul Giangrande g a Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, IRCCS Cà Granda Foundation Maggiore Policlinico Hospital, Milan, Italy b Gulf States Hemophilia and Thrombophilia Center, University of Texas Health Science Center, Houston, TX, USA c IHTC, Cláudio L.P. Correa, Hemocentro Unicamp, INCT do Sangue, University of Campinas, Campinas, São Paulo, Brazil d Ortho-trauma Unit, Emergency Department, IRCCS Cà Granda Foundation Maggiore Policlinico Hospital, Milan, Italy e Novo Nordisk A/S, Søborg, Denmark f Novo Nordisk Health Care AG, Zürich, Switzerland g Oxford Haemophilia and Thrombosis Centre, Oxford University Hospitals NHS Trust, Oxford, UK ARTICLE INFO ABSTRACT Keywords: Congenital haemophilia with inhibitors On-demand treatment Recombinant activated factor VII Surgery The availability of recombinant activated factor VII (rfviia, eptacog alfa activated) has greatly advanced the care of patients with haemophilia A or B who have developed inhibitors against the infused replacement factor. Recombinant FVIIa is licensed for the on-demand treatment of bleeding episodes and the prevention of bleeding in surgery or invasive procedures in patients with congenital haemophilia with inhibitors. This article attempts to review in detail the extensive evidence of rfviia in congenital haemophilia patients with inhibitors. Patients with acute bleeding episodes are best treated on demand at home, to achieve the short- and long-term benefits of rapid bleed control. Key prospective studies have shown that rfviia achieves consistently high efficacy rates in the management of acute (including joint) bleeds in inhibitor patients in the home treatment setting. Substantial postapproval data from key registries also support the on-demand efficacy profile of rfviia established by the prospective clinical trials. The availability of rfviia has allowed major surgery to become a reality for inhibitor patients. Studies in key surgery, including orthopaedic procedures, have found that rfviia provides consistently high efficacy rates. Importantly, the wealth of data does not raise any unexpected safety concerns surrounding rfviia use; this is likely because rfviia is a recombinant product with a localised mechanism of action at the site of vascular injury. In summary, rfviia is established as an effective and well-tolerated first-line treatment for on-demand bleeding control and bleed prevention during minor and major (including elective orthopaedic) surgery in inhibitor patients. Use of rfviia has been a major step towards narrowing the gap in outcomes between inhibitor patients and noninhibitor patients Elsevier Ltd. All rights reserved. 1. Introduction The most serious and challenging complication of haemophilia treatment is the development of alloantibodies, or inhibitors, against infused factor (F) VIII (FVIII; haemophilia A) or FIX (haemophilia B) [1 3]. Inhibitor development occurs more frequently among patients with severe or moderately severe haemophilia, and is more common in haemophilia A than in haemophilia B; approximately 30% of patients with severe haemophilia A develop inhibitors, while the corresponding figure for severe haemophilia B is 1 6% [1,2]. In haemophilia patients with high-responding inhibitors, standard replacement therapy with FVIII or FIX concentrates is * Corresponding author. address: Hemophilia_ctr@policlinico.mi.it (E. Santagostino). Tel.: ; fax: usually ineffective, resulting in poor control of haemorrhagic episodes [1,4]. This, in turn, increases the risk of morbidity, mortality, orthopaedic complications and disability, as well as reduced quality of life, compared with patients without inhibitors [1,2,5 7]. The ultimate goal of treatment for these patients is eradication of the inhibitor through immune tolerance induction (ITI) [1,2,8], which involves frequent and long-term administration of factor concentrates in an attempt to tolerise the immune system to FVIII or FIX, and thus restore responsiveness to factor replacement therapy [1,2]. However, ITI fails in approximately 30% of inhibitor patients with haemophilia A [2,8], and is not commonly attempted in those with haemophilia B, as re-exposure to FIX may trigger severe anaphylactoid reactions and nephrotic syndrome [2,8]. When ITI fails or is not feasible, the treatment of bleeding episodes, and the prevention or reduction of bleed-related morbidity, remain significant challenges. In patients with transient, X/ 2015 Elsevier Ltd. All rights reserved.

2 S10 E. Santagostino et al. / Blood Reviews 29 S1 (2015) S9 S18 low-responding and/or low-titre (<5 Bethesda units [BU]/mL) inhibitors, increased doses of FVIII or FIX may be sufficient to overcome the inhibitor and provide haemostasis [1,2,8]. In most patients with high-responding and high-titre (>5 BU/mL) inhibitors, however, treatment with bypassing agents is required [1,8]. The two bypassing agents currently licensed for use in inhibitor patients are recombinant activated factor VII (rfviia, eptacog alfa activated, NovoSeven, Novo Nordisk A/S, Bagsvaerd, Denmark) and plasma-derived activated prothrombin complex concentrate (pd-apcc, Factor Eight Inhibitor Bypassing Activity, FEIBA, Baxter, Deerfield, IL, USA). These agents are able to manage bleeding in the presence of inhibitors; they do not attempt to restore the normal pathways of haemostasis, but instead, boost thrombin generation despite a lack of plateletsurface FVIIIa FIXa ( tenase ) activity [9,10]. The availability of bypassing agents has greatly advanced the care of haemophilia patients with inhibitors [4]. In the UK, for example, the mortality rate for these patients was twice that observed in patients without inhibitors between 1977 and 1992; however, the mortality rate was similar for patients with and without inhibitors between 1993 and 1999 a period that corresponded with increased use of pd-apcc, increased popularity of ITI and the introduction of rfviia [6]. During this time, rfviia was seen as particularly innovative, as it was the first and only recombinant FVII product available for inhibitor patients, that also targets the site of injury. Therefore, it was predicted to provide inhibitor patients with a safe and effective haemostatic treatment. From 1988 to 1999, rfviia was available on an emergency and compassionate use basis at sites in Europe and North America, and early experience revealed it to be a major breakthrough in the treatment of haemophilia patients with inhibitors, as it provided life-saving therapy in life-threatening situations. In an open-label, uncontrolled, emergency-use study, for example, rfviia was well tolerated and effectively controlled life-threatening intracranial haemorrhage in 10/11 patients with haemophilia who were previously unresponsive to alternative therapies [11]. The benefits of improved care are not yet fully realised worldwide, however: economic constraints in developing countries mean that many patients in these regions receive no, or minimally adequate, treatment, resulting in a greater burden of disability, pain and early death [12,13]. Recombinant FVIIa is currently licensed for the on-demand treatment of bleeding episodes and for the prevention of bleeding in surgery or invasive procedures in patients with congenital haemophilia and inhibitors [14]. It was first approved in Europe in 1996 at a labelled dose of 90 μg/kg every 2 3 hours, and in the United States in 1999 at 90 μg/kg every 2 hours [14,15]. More recently, a single dose regimen of 270 μg/kg was approved by the European Medicines Agency (EMA) in March 2007; this was followed by other approvals worldwide [14]. Data on the pharmacokinetics (PK) and safety of rfviia at a single dose of 270 μg/kg show that it has a PK profile in line with what can be expected in comparison to that observed for the standard 90 μg/kg dose [16], and that it also has the same tolerability, safety and dose-dependent activity profiles established by the 90 μg/kg dose [16]. In this article, we review the wealth of data on the clinical efficacy of rfviia in inhibitor patients, paying particular attention to on-demand therapy for acute bleeds and prevention of surgical bleeds. 2. On-demand therapy for acute bleeds 2.1. The importance of rapid bleeding control A growing body of evidence suggests that the benefits of rapid bleed control include fast pain relief [17 19], early restoration of joint mobility [19] and reduced hospitalisation [20,21]. Together, these immediate benefits of rapid bleed control can decrease joint damage in the short term and may translate into long-term reductions in arthropathy, disability and the need for orthopaedic intervention, with subsequent improvements in quality of life [5]. Timely administration of treatment is crucial to achieving rapid bleed control; clinical guidelines, including those from the World Federation of Hemophilia [22], recommend treating acute bleeds within 2 hours of onset to ensure favourable outcomes in all patients with haemophilia. Home treatment is widely considered to be the best way of avoiding treatment delays [5] Prospective studies of rfviia in home treatment Five key prospective studies (including two phase III) demonstrated that rfviia achieves consistently high efficacy rates when used to manage acute (including joint) bleeds in haemophilia patients with inhibitors in the home treatment setting (Table 1). These data provided an important step towards the goal of providing inhibitor patients with treatment outcomes matching those of non-inhibitor patients. The first phase III trial was a multicentre, open-label, 1-year, single-arm study in which 56 patients were administered up to three doses of rfviia (90 μg/kg) at 3-hourly intervals within 8 hours of the onset of a mild or moderate bleed [23]. Once the subject considered that rfviia had been effective with regard to haemostasis (after 1 3 doses), one further (maintenance) dose was administered. Haemostasis was rated effective in 566/614 (92%) evaluable bleeds; mean time to treatment initiation for successfully treated cases was 1.1 hours, and a mean of 2.2 injections was required for haemostasis. These findings led to the conclusion that early administration of rfviia achieves haemostasis after 1 3 doses [23]. Furthermore, efficacy (in terms of both achieving and maintaining haemostasis) was similar for all joint, target joint and muscle bleeds, and outcomes were comparable for mild and moderate bleeds [23]. The second phase III trial, representing the largest of its kind ever conducted in patients with congenital haemophilia A or B with inhibitors, had an international, randomised, double-blind, cross-over design and compared vatreptacog alfa (a modified rfviia molecule) with a standard rfviia regimen ( μg/kg) in the treatment of all bleeds, including joint bleeds (Table 1) [24]. In total, 567 bleeds in 69 patients were treated. Of these, 227 bleeds in 57 patients were treated with rfviia, which achieved an efficacy rate of 93% at 12 hours. Median time to treatment initiation was 0.4 hours (mean, 1.3 hours). Importantly, rfviia provided rapid bleed control: late-treated bleeds (>2 hours) responded as well to treatment as early-treated bleeds ( 2 hours) when evaluating the need for additional haemostatic medication after 12 hours; early-treated bleeds had a better response when evaluated at 24 or 48 hours. These results confirmed the wellestablished efficacy of rfviia since its market authorisation 20 years ago [24]. Three randomised crossover trials demonstrated comparable efficacy between a single high dose of rfviia and standard-dose rfviia in the treatment of joint bleeds in the home treatment setting. In the first, 18 inhibitor patients were randomised to a standard rfviia dosing regimen of 90 μg/kg every 3 hours or a single dose of 270 μg/kg, and were instructed to treat four consecutive joint bleeds (ankles, knees or elbows) within 6 hours of onset [25]. Treatment success ( effective treatment and a visual analogue scale [VAS] score 70) over 48 hours was similar between the two dosing regimens (31 66% for the standard dose; 25 64% for the high dose), although the number of infusions needed was significantly higher for the standard versus high dose regimen (Table 1). A partial response ( partially effective

3 E. Santagostino et al. / Blood Reviews 29 S1 (2015) S9 S18 S11 Table 1 Trial designs of key prospective studies of recombinant activated FVII in the home treatment of bleeds (including joint bleeds) in haemophilia patients with inhibitors Trial design No. of patients No. of bleeds treated with rfviia Bleed type Dosing regimen No. of injections Key et al., 1998 [23] Multicentre, open-label, single arm, phase III trial (1 year) Mild to moderate joint, muscle and mucocutaneous bleeds 90 μg/kg at 3-hour intervals within 8 hours of bleed onset ( 3 doses), followed by one maintenance dose once haemostasis has been achieved Mean, 2.2 Lentz et al., 2014 [24] Multicentre, randomised, double-blind, active-controlled, cross-over, phase III trial All bleeds, irrespective of location or severity μg/kg Mean, 2.52 Santagostino et al., 2006 [25] Multicentre, randomised, open-label, cross-over, prospective trial standard dose 36 high dose Joint (including target joint) bleeds Standard dose: 90 μg/kg every 3 hours within 6 hours of bleed onset Single high dose: 270 μg/kg within 6 hours of bleed onset Median, 3 and 1 for standard and high dose regimens, respectively* Kavakli et al., 2006 [17] Multicentre, randomised, cross-over, double-blind trial Joint bleeds First bleed treated with a single dose of 270 μg/kg followed by two doses of placebo at 3-hour intervals, then second bleed treated with 3 90 μg/kg at 3-hour intervals, or vice versa Young et al., 2008 [26] Randomised, multicentre, cross-over, double-blind trial 24 patients (1 270 μg/kg) 46 Joint bleeds μg/kg followed by two doses of placebo 22 patients (3 90 μg/kg) 3 90 μg/kg * P = treatment and a VAS score >30 to <70) over 48 hours was also similar between the two dose regimens (31 59% for the standard dose; 22 56% for the high dose). Median rfviia consumption per successful course was identical between the two dosage groups (270 μg/kg) [25]. Kavakli et al. [17] conducted a cross-over trial with a doubleblind design. They randomised 22 patients to treat a first joint bleed with a single dose of rfviia 270 μg/kg followed by two doses of placebo, and then a second joint bleed with rfviia 90 μg/kg 3, or vice versa. Successful bleed control was assessed using a global response tool and by the evaluation of need for an additional dose at 9 hours (similar to Key et al. [23]). Comparable efficacy was achieved with standard and high dose regimens (85.7% and 90.5%, respectively) [17]. In the third crossover trial, by Young et al., 27 patients received each of three treatments in one of six possible sequences for joint bleeds: μg/kg rfviia followed by two doses of placebo; 3 rfviia 90 μg/kg (blinded); and 1 75 IU/kg pd-apcc (open-label) [26]. The efficacy assessment tools were similar to those in the Kavakli et al. study, but the inclusion criteria were different: in Kavakli et al., eligible patients were required to have had three bleeds within the previous 12 months, whereas in Young et al., they were required to have had two bleeds [26]. In addition, the more severely affected patients in the trial by Kavakli et al. may have assessed pain and mobility differently than the less severely affected patients in the trial by Young et al. [26]. In the Young et al. trial, both rfviia regimens showed similar efficacy results (as assessed by the requirement for additional haemostatics within 9 hours and by a novel global response algorithm). The use of rescue medication was 9.1% and 8.3% in the standard and high dose rfviia groups, respectively, whereas 36.4% of patients treated with pd-apcc required rescue medication [26]. The percentage of patients achieving haemostasis without requiring use of rescue medication within 9 hours of the first administration of rfviia was 90.9% and 91.7% for the standard and high dose groups, respectively. The percentage of patients in the 270 μg/kg group who required additional haemostatics within 9 hours (8.3%) was significantly lower than that for the pd-apcc group [26]. Together, the prospective cross-over clinical trials show that a single dose of rfviia 270 μg/kg is at least as effective as rfviia 3 90 μg/kg [17,25,26], does not increase consumption versus standard dosing [25], and may offer improved convenience that leads to increased compliance [17,25] Key registry data on rfviia efficacy Clinical registries and databases play a valuable role in the management of rare disorders, including haemophilia, by providing a source of data for the study of treatment modalities and outcomes [27 29]. Several key registries sponsored by Novo Nordisk have collected substantial post-approval data on the efficacy and safety of rfviia for on-demand treatment in realworld clinical practice (Table 2). Established in 1999 as a joint effort between the Hemophilia Research Society (HRS) and Novo Nordisk Inc., the US-based Hemostasis and Thrombosis Research Society (HTRS) registry [21] was a large-scale database that systematically collected observational patient data up to November 2008 to provide a platform for the post-marketing surveillance of rfviia. The

4 S12 E. Santagostino et al. / Blood Reviews 29 S1 (2015) S9 S18 Table 2 Key registry data on recombinant activated FVII use in the treatment of acute bleeds in haemophilia patients with inhibitors No. of patients/ bleeds Bleed type Dosing regimen No. of injections Efficacy Hemostasis and Thrombosis Research Society registry [21] 38/555 Mild to severe bleeds in all locations (including joints) a Categorised by dose: <100 μg/kg (n = 146) Median total dose/bleed, μg/kg μg/kg (n = 154) Median total dose/bleed, μg/kg μg/kg (n = 136) Median total dose/bleed, μg/kg >200 μg/kg (n = 119) Median total dose/bleed, μg/kg Mean no. of injections: <100 μg/kg: μg/kg: μg/kg: 3.4 >200 μg/kg: 2.3 At 72 hours: b <100 μg/kg: 84.9% μg/kg: 84.4% μg/kg: 83.8% >200 μg/kg: 96.6% ONE registry [30] 85/494 Mild to moderate bleeds in all locations (including target joints) a Categorised by initial dose: LD: 120 μg/kg (n = 156) Median initial dose, 90 μg/kg ID: >120 to <250 μg/kg (n = 127) Median initial dose, 174 μg/kg HD: 250 μg/kg (n = 211) Median initial dose, 270 μg/kg Median (range) no. of injections: LD: 2 (1 17) ID: 2 (1 23) HD: 1 (1 7) At 9 hours: c LD: 85% ID: 96% HD: 86% HemoRec registry [29] 15/363 Mild to moderate bleeds <120 μg/kg (105 bleeds) or 120 μg/kg (258 bleeds) Mean no. of injections: Treatment initiated 2 hours: 1.2* Treatment initiated >2 hours: 2.1 Mean duration of bleeds: All bleeds (n = 363) <120 μg/kg: 11.6 hours 120 μg/kg: 7.8 hours** Treatment initiated 2 hours: 5.6 hours* Treatment initiated >2 hours: 16.3 hours Joint bleeds (n = 288) <120 μg/kg: 9.6 hours 120 μg/kg: 5.6 hours** Treatment initiated 2 hours: 5.6 hours* Treatment initiated >2 hours: 9.9 hours LD, low dose; ID, intermediate dose; HD, high dose. a Includes spontaneous, traumatic and procedure-related bleeds. b Bleeding stopped at 72 hours. c Effective and partially effective haemostasis at 9 hours for all acute bleed episodes. * P < versus treatment initiated >2 hours. ** P < versus doses <120 μg/kg. database represents the most comprehensive collection of data on rfviia use since its US licensure, and supports the efficacy profile of rfviia established by the prospective clinical trials (Table 2) [21]. The ONE registry was an international, prospective, observational study of on-demand rfviia for the treatment of mild to moderate bleeds in inhibitor patients; its primary aim was to compare outcomes associated with single dose rfviia (270 μg/kg) versus those with standard dosing (3 90 μg/kg; Table 2) [30]. The registry was initiated in March 2008 and involved 42 centres in 16 countries; data collection was completed in July Overall, it revealed that fewer injections were required for bleeds treated with initial high doses, but that single and standard dose treatments were associated with similar overall effectiveness (Table 2) [30]. HemoRec is a prospective, observational, web-based Czech registry designed to capture and analyse a broad range of realworld data on the treatment of haemophilia patients. Data collection commenced in February 2006 and is ongoing. To date, this registry shows that all reported bleeds analysed in haemophilia patients with inhibitors stop with rfviia treatment, with higher initial doses and early treatment initiation achieving significantly faster bleed resolution (Table 2) [29] Safety in on-demand use The safety of rfviia has been well studied; no unexpected safety concerns have been raised by any of the key prospective trials [17,23 26,31] or registry data [21,29 31] that have assessed on-demand rfviia in acute bleed treatment. Importantly, no thromboembolic events (TEs) or neutralising antibodies have been reported in rfviia-treated patients with congenital haemophilia. The favourable safety profile of rfviia is a result of it being a recombinant product (carrying no risk of transferring bloodborne pathogens) and having a localised mechanism of action at the site of vascular injury. Unlike pd-apcc, in which small amounts of FVIII may be found, because rfviia is a recombinant product, it has no potential for anamnesis [32]. There are also

5 E. Santagostino et al. / Blood Reviews 29 S1 (2015) S9 S18 S13 Table 3 Key studies of secondary prophylaxis with recombinant activated factor VII Trial design No. of patients Dosing regimen Weekly dose Efficacy Konkle et al., 2007 [35] Multicentre, randomised, double-blind, parallel-group trial μg/kg or 270 μg/kg once daily for 3 months 90 μg/kg: 45% reduction in estimated mean bleeding frequency (from 5.6 to 3 bleeds/month)* 43% reduction in target joint bleeds* 270 g/kg: 59% reduction in estimated mean bleeding frequency (from 5.3 to 2.2 bleeds/month)* 61% reduction in target joint bleeds** Young et al., 2012 [38] Retrospective, observational study (PRO-PACT) 74 total ( 1 bleed during the pre-prophylaxis period) Subset of 36 frequent bleeders ( 1 bleed/ month during the preprophylaxis period) Median, g/kg (1 14 times/week) Median, g/kg Total population: 46% reduction in estimated mean bleeding frequency (from 1.37 to 0.74 bleeds/ month) 40% reduction in joint bleeds Frequent bleeders: 52% reduction in estimated mean bleeding frequency (from 2.1 to 1.01 bleeds/month) 46% reduction in joint bleeds * P ** P < no maximum daily dose restrictions for rfviia. The article by Neufeld et al., in this supplement [33], summarises the safety profile of rfviia treatment. 3. Secondary prophylaxis in inhibitor patients Although rfviia is not currently licensed worldwide for prophylaxis, the past decade has witnessed a growing number of reports on its use as secondary prophylaxis in inhibitor patients [34 38]. Only one prospective, randomised, double-blind trial has explored the efficacy and safety of secondary prophylaxis with rfviia in inhibitor patients with frequent bleeds [35]. Twenty-two patients received a blinded dose of rfviia 90 μg/kg or 270 μg/kg once daily for 3 months, followed by a 3-month post-prophylaxis period. Both dose regimens significantly reduced the estimated mean bleeding frequency during prophylaxis (reductions of 45% and 59% for 90 μg/kg and 270 μg/kg, respectively; not statistically different from each other) compared with on-demand therapy (Table 3); these reductions were maintained during the postprophylaxis period. There were no significant differences in efficacy between the two doses, and the efficacy of prophylaxis appeared to be most pronounced for spontaneous joint bleeds. The authors concluded that the trial provides evidence for the concept of secondary rfviia prophylaxis in inhibitor patients with frequent bleeds (four or more bleeds per month) [35]. A more recent, observational, retrospective study (PRO-PACT), representing the largest real-world data collection for rfviia use in prophylaxis, described clinical experience of secondary prophylaxis with rfviia [38]. Data were collected from medical records to evaluate outcomes such as dosing regimens and bleeding events. In accordance with the prospective trial, this study found that secondary prophylaxis with rfviia reduced estimated mean bleeding frequency relative to the pre-prophylaxis period; reductions were similar in the total population (46%) and a sub-population of patients with more frequent bleeds prior to prophylaxis (52%; Table 3), although a dose response was not detected [38]. The study also reported a decreased frequency of healthcare resource consumption among patients on prophylaxis (including hospital admissions and school/work absenteeism), indicating potential quality-of-life benefits [38]. There were no reports of TE events or any new safety concerns in either of the two studies [35,38]. 4. Bleed prevention during surgery The availability of rfviia has allowed major surgery to become a reality for haemophilia patients with inhibitors. Based on evidence from prospective clinical trials [39,40], rfviia is indicated for the prevention of bleeding in haemophilia patients with inhibitors who are undergoing surgery or invasive procedures [14]. In Europe, the Summary of Product Characteristics states that rfviia should be administered pre-operatively at 90 μg/kg (repeated after 2 hours), at 90 μg/kg every 2 3 hours for 1 2 days immediately post-operatively, and at 90 μg/kg every 2 4 hours for 6 7 days following surgery, after which the dose interval may be increased to 6 8 hours for a further 2 weeks. However, a range of rfviia dosing regimens has been used for surgical procedures [41] (other helpful information on surgery are available in the reviews by Rodriguez-Merchan, 2012 [42] and Valentino et al., 2011 [43]). Current treatment recommendations and guidelines for surgery are summarised in Table Elective orthopaedic surgery It is well established that inhibitor patients have a worse orthopaedic status than non-inhibitor patients, with subsequent increases in psychosocial difficulties and reductions in quality of life [7,41,44]. Orthopaedic surgery is particularly challenging in the presence of inhibitors; before the introduction of bypassing agents, elective surgery in inhibitor patients was not feasible, and before the advent of rfviia, it was uncommon [4,41,44 46]. The availability of bypassing agents has thus transformed surgical practice for the inhibitor population, and rfviia has allowed many inhibitor patients to undergo elective orthopaedic surgery (EOS) safely and with a high degree of efficacy [44 47]. The first pioneering use of rfviia in humans was in EOS, due to the unmet clinical need for an effective haemostatic medication in this indication [48]. In a recent survey of US haemophilia treatment centres (HTCs), 42/85 participating centres reported that they had performed 455 orthopaedic procedures in inhibitor patients

6 S14 E. Santagostino et al. / Blood Reviews 29 S1 (2015) S9 S18 Table 4 Key recommendations and guidelines for recombinant activated factor VII use in surgery Pre-operative dosing Immediate post-operative dosing Days/weeks after surgery Mode of administration Mathew et al., 2007 [57] μg/kg μg/kg every 2 hours for 48 hours μg/kg every 4 hours from day 3 to 4, then every 6 hours for another week Bolus infusion Continuous infusion used only as part of a clinical trial a Obergfell et al., 2008 [58] Minimum initial bolus of 120 μg/kg μg/kg every 2 hours (bolus) 50 μg/kg/hour (continuous infusion) a Bolus or continuous infusion Teitel et al., 2009 [59] 90 μg/kg immediately pre-operatively 90 μg/kg every 2 hours for 48 hours Dose interval can be increased based on clinical outcome (grade 1A recommendation) Bolus infusion Continuous infusion after initial bolus appears promising (grade 2B) but is not recommended for routine use Giangrande et al., 2009 [47] μg/kg immediately prior to first incision, then 90 μg/kg every 2 hours throughout surgery b 90 μg/kg every 2 hours for 90 μg/kg every 3 hours for 48 hours b 48 hours if haemostasis is good 90 μg/kg every 4 hours for 3 days 90 μg/kg every 6 hours from day 8 until discharge b Bolus infusion Escobar et al., 2012 [41] μg/kg pre-incision, followed by μg/kg every 2 hours for the first 48 hours If haemostasis is adequate, can start increasing the interval to μg/kg every 3 hours for 48 hours, μg/kg every 4 hours for 3 days, μg/kg every 6 hours from day 8 until discharge a Continuous infusion not included in Summary of Product Characteristics for NovoSeven. b Concomitant administration of oral tranexamic acid is recommended (25 mg/kg every 6 8 hours, starting the evening before surgery and continuing until discharge). during the previous decade [49]. When considering all types of surgical procedures, rfviia was the agent most frequently used to provide haemostatic cover, and was commonly used from the pre-operative period through to hospital discharge [49]. Three key studies, sponsored by Novo Nordisk, have demonstrated the efficacy of rfviia in EOS (Table 5). The two earliest were randomised trials, the first of which (published in 1998) was a prospective, double-blind study comparing two doses of rfviia (35 μg/kg and 90 μg/kg) administered for 5 days to cover 10 orthopaedic procedures in 10 patients [40]. In both dose regimens, intra-operative haemostasis was achieved in 100% of patients. Post-operatively (between wound closure and 48 hours), satisfactory haemostasis was observed in 100% of patients receiving 90 μg/kg rfviia and 60% of those receiving 35 μg/kg (Table 5). The study s authors concluded that 90 μg/kg rfviia was effective as a first-line option for surgery in inhibitor patients [40]. The second randomised trial, which had an openlabel design, compared bolus infusions of rfviia with continuous infusion to cover 22 major surgeries, including 13 orthopaedic procedures. Efficacy in the bolus and continuous infusion arms was comparable at post-operative day 10 (73% and 75%, respectively; Table 5), although no separate data were reported for orthopaedic procedures [39]. A recently completed Japanese post-marketing surveillance study assessed rfviia efficacy in surgical and invasive procedures (including 24 orthopaedic procedures) in 29 inhibitor patients [50]. In this study, rfviia effectively prevented surgical bleeding in 89% and 77% of patients with haemophilia A and B, respectively (Table 5). However, while the study provided efficacy data for all surgical procedures, it did not analyse orthopaedic procedures separately [50]; nevertheless, the results from this postmarketing surveillance study support those of the clinical trials by demonstrating consistently high efficacy rates when rfviia is used to cover surgery (including orthopaedic procedures) in inhibitor patients. The initial cost of orthopaedic surgery performed under rfviia may be high, but it has been reported that such surgery is cost-effective in the long term, due to reduced product use and improved quality of life in the months and years following surgery [51] Non-orthopaedic surgery As the haemophilia population gets older, age-related pathologies will become more prevalent; non-orthopaedic surgical procedures may therefore be needed more often [41]. Growing experience of performing surgery in inhibitor patients means that this population has easier access to routine interventions than it did in the recent past. According to a recent survey of US HTCs, the most common non-orthopaedic procedures carried out in inhibitor patients over the last 10 years are central line insertions (73/85 centres) and dental procedures (58 centres), followed by abdominal, cardiovascular and otolaryngological interventions (23, 14 and 11 centres, respectively). These findings indicate that many HTCs are performing surgery in inhibitor patients; in addition, the survey results suggest that rfviia is used more frequently than any other treatment modality to facilitate these procedures [49]. The three studies reporting rfviia use in orthopaedic surgery in inhibitor patients (see section 4.1) also assessed its use in non-orthopaedic procedures. Table 6 summarises these studies along with a US post-marketing surveillance study from the HTRS registry [52]. Most studies did not differentiate between orthopaedic and non-orthopaedic procedures in their efficacy assessments (as discussed in section 4.1); however, one of the randomised trials and the US post-marketing surveillance study did evaluate efficacy according to procedure type. In the prospective randomised trial assessing two doses of rfviia (35 μg/kg and 90 μg/kg) in surgery, 19 non-orthopaedic procedures (18 minor, one major) were recorded [40]. Separate efficacy data were given for the minor procedures, and demonstrated that both intra-operative and post-operative haemostasis were high for both dose groups (Table 6). In the

7 E. Santagostino et al. / Blood Reviews 29 S1 (2015) S9 S18 S15 Table 5 Key data on recombinant activated factor VII use in orthopaedic surgery in congenital haemophilia with inhibitors Type of study/registry No. of orthopaedic procedures Type of procedure (no. of procedures) Dosing regimen Efficacy rate Shapiro et al., 1998 [40] Prospective, doubleblind, randomised, multicentre study 10 Synovectomy (5), hip arthroplasty (2), knee joint manipulation (1), cartilage repair (1), bone graft (1) 35 μg/kg or 90 μg/kg immediately prior to incision; intra-operatively every 2 hours, or as needed; every 2 hours for the first 48 hours; and every 2 6 hours for the following 3 days Intra-operative: a 35 μg/kg: 100% 90 μg/kg: 100% Wound closure to 48 hours: b 35 μg/kg: 60% 90 μg/kg: 100% Pruthi et al., 2007 [39] Open-label, randomised, multicentre trial of bolus infusion versus continuous infusion 16 Knee arthroplasty (8), arthroscopic synovectomy (4), arthrotomy, irrigation and debridement (1) Hip arthroplasty (3) Initial bolus of 90 μg/kg, then 90 μg/kg every 2 hours up to day 5, then every 4 hours for days 6 10 (bolus group) or 50 μg/kg/hour up to day 5, then 25 μg/kg/hour for days 6 10 (continuous infusion group) 73% for bolus infusions c 75% for continuous infusion c Takedani et al., 2014 [50] Japanese post-marketing surveillance study 16 in haemophilia A 8 in haemophilia B Arthroscopic synovectomy d Bolus infusions (n = 5): Median, 125 (range, ) μg/kg per injection Median no. of injections, 39 (29 55) 89% in haemophilia A e 77% in haemophilia B e Continuous infusion (n = 6): Median initial infusion rate: 32 (17 60) μg/kg/hour Joint replacement d Bolus infusions (n = 1): Median, 120 μg/kg per injection Median no. of injections, 72 Continuous infusion (n = 2): 32.5 (29 36) μg/kg/hour n, number of procedures. Efficacy definitions: a Defined as blood loss as expected, or less than expected, during surgery. Note that these efficacy rates relate to all major procedures. There were 11 major procedures in total, including 10 orthopaedic procedures and one laparoscopic renal biopsy. b Defined as effective (bleeding stopped or substantially decreased) and partially effective (bleeding reduced but continuing). There were 11 major procedures in total, including 10 orthopaedic procedures and one laparoscopic renal biopsy. c Haemostasis judged effective at time of therapy discontinuation or post-operative day 10 (whichever was earlier). Note that these efficacy rates relate to all (not just orthopaedic) procedures. d Dosing data only provided for the most frequently performed procedures. e Post-operative bleeding stopped completely or reduced considerably. Note that these efficacy rates relate to all (not just orthopaedic) procedures. US post-marketing surveillance study, data were reported for surgical non-dental, dental, and other medical procedures. While the 54 surgical non-dental surgeries almost certainly included some orthopaedic procedures, it is reasonable to assume that non-orthopaedic interventions would also have been included here. The efficacy of rfviia was % across all surgical and dental procedures in the studies shown in Table Safety in surgical use The ability to perform surgery safely in patients with haemophilia with inhibitors has become possible with rfviia coverage, and this is a major step forward. The two randomised clinical trials [39,40] and two post-marketing surveillance studies [50,52] raised no unexpected safety concerns regarding rfviia use in surgery. Importantly, no neutralising antibodies or inhibitors were reported. One patient in each of the clinical trials developed thrombosis (right internal jugular vein in a patient receiving rfviia 35 μg/kg [40], and left popliteal vein and proximal peroneal vein in a patient receiving bolus rfviia infusions to cover total knee arthroplasty [39]). The outcome of the thrombosis was not recorded in the first case, but in the latter, the patient continued to receive rfviia for two additional days with no adverse consequences [39]. 5. Further considerations for rfviia use There are some clinical scenarios in which rfviia may provide the only safe therapy. FIX inhibitors are strongly associated with the development of potentially life-threatening anaphylactoid reactions, particularly in patients with derangement, or complete deletion, of the FIX gene [10,53]. This precludes treatment with any FIX-containing products, including pd-apcc. Furthermore,

8 S16 E. Santagostino et al. / Blood Reviews 29 S1 (2015) S9 S18 Table 6 Key data on recombinant activated factor VII use in non-orthopaedic surgery Type of study/ registry No. of procedures Type of procedure (no. of procedures) Dosing regimen Efficacy rate Shapiro et al., 1998 [40] Prospective, double-blind, randomised, multicentre study 1 major 18 minor Major: Laparoscopic renal biopsy Minor: CVAD placement (8), placement of other venous access devices (9), Port-a-Cath removal (1) 35 μg/kg or 90 μg/kg immediately prior to incision; intra-operatively every 2 hours, or as needed; every 2 hours for the first 48 hours; and every 2 6 hours for the following 3 days Intra-operative: a 35 μg/kg: 100% 90 μg/kg: 88% Wound closure to 48 hours: b 35 μg/kg: 91% 90 μg/kg: 100% Pruthi et al., 2007 [39] Open-label, randomised, multicentre trial of bolus infusions versus continuous infusion 5 major 2 minor Major: Pseudotumour removal (1), inguinal hernia repair (2), tonsillectomy/adenoidectomy (1), craniotomy and tumour removal (1) Minor: Circumcision (1), entropion and ectropion repair (1) Initial bolus of 90 μg/kg, then 90 μg/kg every 2 hours up to day 5, then every 4 hours for days 6 10 (bolus group) or 50 μg/kg/hour up to day 5, then 25 μg/kg/hour for days 6 10 (continuous infusion group) 73% for bolus infusions c 75% for continuous infusion c Young et al., 2012 [52] HTRS registry 54 surgical non-dental 44 dental Not specified Surgical non-dental: Mean, 111 (55 193) μg/kg per infusion 89% surgical non-dental d 93% dental e Median no. of injections, 21 (1 64) Dental: Mean, 122 (60 295) μg/kg per infusion Median no. of injections, 3 (1 35) Takedani et al., 2014 [50] Japanese postmarketing surveillance study 22 in haemophilia A 5 in haemophilia B Haemophilia A: CVAD/port placement (6), dental/oral (6), gastrointestinal (4), suture/ incision (4), thoracic (2) Haemophilia B: CVAD/port placement (2), dental/oral (1), gastrointestinal (1), suture/ incision (1) Port placement: f Bolus infusions: Median, 109 μg/kg (range, ) per injection Median no. of injections, 15 (3 64) Continuous infusion: Initial infusion rate, 28 μg/kg/hour 89% in haemophilia A g 77% in haemophilia B g CVAD, central venous access device; HTRS, Haemophilia and Thrombosis Research Society. Efficacy definitions: a Defined as surgical blood loss as expected, or less than expected. Note that these efficacy rates relate to all (not just non-orthopaedic) procedures. Data are for the 18 minor bleeds only. b Defined as effective (bleeding stopped or substantially decreased) and partially effective (bleeding reduced but continuing). Data are for the 18 minor bleeds only. c Haemostasis judged effective at time of discontinuation or post-operative day 10 (whichever was earlier). Note that these efficacy rates relate to all (not just non-orthopaedic) procedures. d Bleeding stopped. This efficacy rate covers all surgical non-dental bleeds. e Bleeding stopped. f Dosing data only provided for the most frequently performed procedures. g Bleeding stopped completely or reduced considerably. Note that these efficacy rates relate to all (not just non-orthopaedic) procedures. ITI is rarely attempted in patients with FIX inhibitors because of its association with the additional risk of nephrotic syndrome [10,53]. For these reasons, rfviia may be the only treatment option available for haemophilia B patients with inhibitors, especially for those at high risk of anaphylactoid reactions [10,26,53,54]. While patients with FVIII inhibitors are waiting for their inhibitor titres to decrease before they can begin ITI [3], it is important to prevent exposure to FVIII, as this could trigger an anamnestic response [55]. As pd-apcc contains residual FVIII antigen, its use during this waiting period should be avoided. In contrast, rfviia is devoid of FVIII protein, and is therefore recommended for the treatment of acute bleeds and to cover line insertion in patients waiting for ITI [3,55,56]. 6. Concluding remarks Recombinant FVIIa has played an instrumental role in improving the lives of haemophilia patients with inhibitors by offering the first recombinant molecule with a targeted mechanism of action. Prior to its availability, treatment options for these patients were scarce, outcomes often bleak, and the feasibility of performing orthopaedic surgery almost nonexistent. Importantly, the gap in outcomes between inhibitor patients and non-inhibitor patients is steadily narrowing, and the outlook for inhibitor patients is now far brighter than it was three decades ago. A large, and growing, body of evidence shows that rfviia provides consistently high levels of haemostatic efficacy in both

9 E. Santagostino et al. / Blood Reviews 29 S1 (2015) S9 S18 S17 paediatric and adult patients with haemophilia and inhibitors. Favourable safety and tolerability profiles have also been demonstrated in these indications, even when higher doses are used. Today, rfviia is established as an effective first-line treatment for on-demand bleeding control and bleed prevention during minor and major (including elective orthopaedic) surgery in inhibitor patients. Conflict of interest statement E Santagostino has received research funding from Pfizer and consultancy and speaker fees from Pfizer, Novo Nordisk, Baxter, Bayer, CSL Behring, Biogen Idec/Sobi, Grifols, Octapharma, Kedrion and Roche. M Escobar has received advisory board fees from Novo Nordisk, Baxter, Pfizer, Bayer and CSL Behring. M Ozelo has received research funding from Novo Nordisk, Bayer and CSL Behring, and consultancy and speaker fees from Novo Nordisk, Baxter and Bayer. L Solimeno has received speaker fees from Novo Nordisk, Baxter and Bayer. P Arkhammar is an employee of Novo Nordisk A/S, Denmark. P Giangrande has received speaker and advisory board fees from Novo Nordisk as well as support to attend international conferences. HY Lee and G Rosu are employees of Novo Nordisk Health Care AG, Switzerland. Acknowledgements Sharon Eastwood of PAREXEL, a medical writer supported by funding from Novo Nordisk Health Care AG, provided drafts and editorial assistance to the authors during preparation of this manuscript. References [1] Coppola A, Di Capua M, Di Minno MN, Di Palo M, Marrone E, Leranò P, et al. Treatment of hemophilia: a review of current advances and ongoing issues. J Blood Med 2010;1: [2] DiMichele D. Inhibitors in Hemophilia: a primer. 4th edition. Montréal, Québec: World Federation of Hemophilia; [3] DiMichele DM, Hoots WK, Pipe SW, Rivard GE, Santagostino E. International workshop on immune tolerance induction: consensus recommendations. Haemophilia 2007;13(Suppl. 1):1-22. [4] Shapiro AD, Hedner U. Advances in bypassing agent therapy for hemophilia patients with inhibitors to close care gaps and improve outcomes. Ther Adv Drug Saf 2011;2: [5] Berntorp E. Importance of rapid bleeding control in haemophilia complicated by inhibitors. Haemophilia 2011;17:11-6. [6] Darby SC, Keeling DM, Spooner RJ, Wan Kan S, Giangrande PL, Collins PW, et al. The incidence of factor VIII and factor IX inhibitors in the hemophilia population of the UK and their effect on subsequent mortality, J Thromb Haemost 2004;2: [7] Morfini M, Haya S, Tagariello G, Pollmann H, Quintana M, Siegmund B, et al. European study on orthopaedic status of haemophilia patients with inhibitors. Haemophilia 2007;13: [8] Franchini M, Coppola A, Tagliaferri A, Lippi G. FEIBA versus NovoSeven in hemophilia patients with inhibitors. Semin Thromb Hemost 2013;39: [9] Hoffman M, Dargaud Y. Mechanisms and monitoring of bypassing agent therapy. J Thromb Haemost 2012;10: [10] Mehta R, Parameswaran R, Shapiro AD. An overview of the history, clinical practice concerns, comparative studies and strategies to optimize therapy of bypassing agents. Haemophilia 2006;12(Suppl. 6): [11] Arkin S, Cooper HA, Hutter JJ, Miller S, Schmidt ML, Seibel NL, et al. Activated recombinant human coagulation factor VII therapy for intracranial hemorrhage in patients with hemophilia A or B with inhibitors. Results of the NovoSeven emergency-use program. Haemostasis 1998;28:93-8. [12] Bolton-Maggs PH. Optimal haemophilia care versus the reality. Br J Haematol 2006;132: [13] Skinner MW. WFH: closing the global gap achieving optimal care. Haemophilia 2012;18(Suppl. 4):1-12. [14] NovoSeven Summary of Product Characteristics. europa.eu/docs/en_gb/document_library/epar_-_product_information/ human/000074/wc pdf. [Accessed March 3, 2015]. [15] NovoSeven Highlights of Prescribing Information. downloads/.../ucm pdf. [Accessed March 3, 2015]. [16] Morfini M, Bjerre J. Pharmacokinetics and safety of a 270 mcg kg -1 dose of room temperature stable recombinant activated factor VII in patients with haemophilia. Haemophilia 2011;17: [17] Kavakli K, Makris M, Zulfikar B, Erhardtsen E, Abrams ZS, Kenet G, et al. Home treatment of haemarthroses using a single dose regimen of recombinant activated factor VII in patients with haemophilia and inhibitors. A multi-centre, randomised, double-blind, cross-over trial. Thromb Haemost 2006;95: [18] Kenet G, Lubetsky A, Luboshitz J, Martinowitz U. A new approach to treatment of bleeding episodes in young hemophilia patients: a single bolus megadose of recombinant activated factor VII (NovoSeven). J Thromb Haemost 2003;1: [19] Santagostino E, Gringeri A, Mannucci PM. Home treatment with recombinant activated factor VII in patients with factor VIII inhibitors: the advantages of early intervention. Br J Haematol 1999;104:22-6. [20] Lusher JM. Early treatment with recombinant factor VIIa results in greater efficacy with less product. Eur J Haematol Suppl 1998;63:7-10. [21] Parameswaran R, Shapiro AD, Gill JC, Kessler CM, HTRS Registry Investigators. Dose effect and efficacy of rfviia in the treatment of haemophilia patients with inhibitors: analysis from the Hemophilia and Thrombosis Research Society Registry. Haemophilia 2005;11: [22] Srivastava A, Brewer AK, Mauser-Bunschoten EP, Key NS, Kitchen S, Llinas A, et al. Guidelines for the management of hemophilia. Haemophilia 2013; 19:e1-47. [23] Key NS, Aledort LM, Beardsley D, Cooper HA, Davignon G, Ewenstein BM, et al. Home treatment of mild to moderate bleeding episodes using recombinant factor VIIa (NovoSeven) in haemophiliacs with inhibitors. Thromb Haemost 1998;80: [24] Lentz SR, Ehrenforth S, Karim FA, Matsushita T, Weldingh KN, Windyga J, et al. Recombinant factor VIIa analog in the management of hemophilia with inhibitors: results from a multicenter, randomized, controlled trial of vatreptacog alfa. J Thromb Haemost 2014;12: [25] Santagostino E, Mancuso ME, Rocino A, Mancuso G, Scaraggi F, Mannucci PM. A prospective randomized trial of high and standard dosages of recombinant factor VIIa for treatment of hemarthroses in hemophiliacs with inhibitors. J Thromb Haemost 2006;4: [26] Young G, Shafer FE, Rojas P, Seremetis S. Single 270 microg kg(-1)-dose rfviia vs. standard 90 microg kg(-1)-dose rfviia and APCC for home treatment of joint bleeds in haemophilia patients with inhibitors: a randomized comparison. Haemophilia 2008;14: [27] Dolan G, Makris M, Bolton-Maggs PH, Rowell JA. Enhancing haemophilia care through registries. Haemophilia 2014;20(Suppl. 4): [28] Salaj P, Brabec P, Penka M, Pohlreichova V, Smejkal P, Cetkovsky P, et al. Effect of rfviia dose and time to treatment on patients with haemophilia and inhibitors: analysis of HemoRec registry data from the Czech Republic. Haemophilia 2009;15: [29] Salaj P, Ovesna P, Penka M, Hedner U. Analyses of recombinant activated factor VII treatments from clinical practice for rapid bleeding and acute pain control in haemophilia patients with inhibitors. Haemophilia 2012;18:e [30] Chambost H, Santagostino E, Laffan M, Kavakli K, ONE Registry Steering Committee on behalf of the investigators. Real-world outcomes with recombinant factor VIIa treatment of acute bleeds in haemophilia patients with inhibitors: results from the international ONE registry. Haemophilia 2013;19: [31] Shapiro AD, Neufeld EJ, Blanchette V, Salaj P, Gut RZ, Cooper DL. Safety of recombinant activated factor VII (rfviia) in patients with congenital haemophilia with inhibitors: overall rfviia exposure and intervals following high (>240 g kg -1 ) rfviia doses across clinical trials and registries. Haemophilia 2014;20:e [32] Kempton CL, White GC. How we treat a hemophilia A patient with a factor VIII inhibitor. Blood 2009;113:11-7. [33] Neufeld EJ, Négrier C, Arkhammar P, Benchikh el Fegoun S, Simonsen MD, Rosholm A, et al. Safety update on the use of recombinant activated factor VII in approved indications. Blood Reviews 2015;29(Suppl. 1):S [34] Franchini M, Manzato F, Salvagno GL, Montagnana M, Zaffanello M, Lippi G. Prophylaxis in congenital hemophilia with inhibitors: the role of recombinant activated factor VII. Semin Thromb Hemost 2009;35: [35] Konkle BA, Ebbesen LS, Erhardtsen E, Bianco RP, Lissitchkov T, Rusen L, et al. Randomized, prospective clinical trial of recombinant factor VIIa for secondary prophylaxis in hemophilia patients with inhibitors. J Thromb Haemost 2007;5: [36] Morfini M, Auerswald G, Kobelt RA, Rivolta GF, Rodriguez-Martorell J, Scaraggi FA, et al. Prophylactic treatment of haemophilia patients with inhibitors: clinical experience with recombinant factor VIIa in European Haemophilia Centres. Haemophilia 2007;13: [37] Young G, McDaniel M, Nugent DJ. Prophylactic recombinant factor VIIa in haemophilia patients with inhibitors. Haemophilia 2005;11: [38] Young G, Auerswald G, Jimenez-Yuste V, Lambert T, Morfini M, Santagostino E, et al. PRO-PACT: retrospective observational study on the prophylactic use of recombinant factor VIIa in hemophilia patients with inhibitors. Thromb Res 2012;130: