The first recombinant FVIII produced in human cells an update on its clinical development programme

Haemophilia (2014), 20 (Suppl. 1), 1 9 DOI: 10.1111/hae.12322 REVIEW ARTICLE The first recombinant FVIII produced in human cells an update on its clinical development programme L. A. VALENTINO,* C. NEGRIER, G. KOHLA, A. TIEDE, R. LIESNER, D. HART** and S. KNAUB *Departments of Pediatrics, Internal Medicine, Immunology/Microbiology and Biochemistry, Section of Pediatric Hematology/ Oncology, RUSH Hemophilia and Thrombophilia Center, RUSH University Medical Center, Chicago, IL, USA; Hematology Division, Hemophilia Comprehensive Care Center, Hopital Edouard Herriot Pavillon E, Universite Lyon 1, Lyon, France; Octapharma R&D Molecular Biochemistry Berlin, Berlin, Germany; Hematology, Hemostasis, Oncology and Stem Cell Transplantation, Hannover Medical School, Hannover, Germany; Paediatric Haemostasis and Thrombosis, Haemophilia Comprehensive Care Centre, Great Ormond Street Hospital for Children NHS Trust, London, UK; **Haemostasis, Blizard Institute and Genome Centre, Barts and the London School of Medicine and Dentistry, The Royal London Hospital, London, UK; and Clinical R&D, Haematology Octapharma AG, Lachen, Switzerland Summary. The development of inhibitors and the need for frequent venous access for FVIII injection are major challenges in current haemophilia treatment. Presently available recombinant FVIII (rfviii) products produced in hamster cell lines are associated with inhibitor formation in up to 32% of previously untreated patients. The new human cell line-derived recombinant human FVIII (Human-cl rhfviii) protein is the first native, unmodified truly human rfviii product produced in a human cell line without additive animal proteins. The aim of using a human cell line for the production of rfviii is the avoidance of non-human epitopes on rfviii, thereby potentially reducing the rate of inhibitor development, avoiding allergic reactions and allowing personalized prophylaxis Introduction with the chance of fewer infusions. Studies to date show that prophylaxis with Human-cl rhfviii prevents 96% of bleeding events in adults with severe haemophilia A when compared to on-demand treatment. Available pharmacokinetic data with a mean half-life of 17.1 h allow personalized prophylaxis with the chance of fewer infusions. Studies in previously treated children and adults indicate that Human-cl rhfviii is efficacious and safe in the prevention and treatment of bleeding episodes and that none of the treated patients developed inhibitors or allergic reactions thus far. Keywords: efficacy, haemophilia A, human recombinant FVIII, immunogenicity, pharmacokinetics, safety Inhibitor formation and the need for frequent venous access for factor VIII (FVIII) injection are major challenges in current haemophilia treatment. Presently available recombinant FVIII (rfviii) products produced in hamster cell lines are associated with a cumulative incidence of inhibitors in up to 32% of Correspondence: Prof. Leonard A. Valentino MD, Departments of Pediatrics, Internal Medicine, Immunology/Microbiology and Biochemistry, Director, Section of Pediatric Hematology/Oncology Director, RUSH Hemophilia and Thrombophilia Center, RUSH University Medical Center, 1653 West Congress Parkway, Chicago, IL 60612-3833, USA. Tel.: +1-312-942-8114; fax: +1-312-942-8975; E-mail: lvalentino@rush.edu Accepted after revision 22 October 2013 previously untreated patients (PUPs). The new human cell line-derived recombinant human FVIII (Human-cl rhfviii) protein is the first native, unmodified human rfviii product produced in a human cell line. Human-cl rhfviii is manufactured without additive animal- or human-derived materials during production and purification and consists of a heavy chain followed by a 16 amino acid linker sequence and a light chain. Due to its human cell origin, Human-cl rhfviii does not carry antigenic non-human epitopes and has thus the potential to satisfy the unmet needs of the global haemophilia community by reducing the rate of inhibitors, avoiding allergic reactions and allowing personalized prophylaxis with fewer infusions. In contrast to currently available commercial rfviii produced in hamster cell lines, Human-cl rhfviii is fully sulphated at Tyr 1680 and thus displays higher binding affinity for von Willebrand factor (VWF) than 2014 John Wiley & Sons Ltd 1

2 L. A. VALENTINO et al. current rfviii products. A higher binding affinity to VWF minimizes the circulation of unbound FVIII and reduces FVIII clearance. Furthermore, similar FVIII activities in one-stage and chromogenic assays have been observed with Human-cl rhfviii indicating that monitoring with either assay is equally valid. The results of the recently completed Phase II/III multicenter clinical trials in severe haemophilia A will be presented. Furthermore, an insight into the new personalized prophylactic study NuPreviq and the ongoing PUP study (NuProtect) will be provided. Human-cl rhfviii: functional insights gleaned from structure Twenty years after the introduction of rfviii derived from hamster cell lines there are still opportunities for improvements: 1. Reduce the overall immunogenic challenge. 2. Provide full functional properties, for example high-affinity binding of FVIII towards VWF. 3. Increase tolerability. 4. Set an even higher level of (theoretical) pathogen safety. There are various risk factors for the development of inhibitors; these include genetic, non-genetic and FVIII product-related risk factors, the latter includes the primary structure of the product, for example single nucleotide polymorphisms, aggregation (formulation) and posttranslational modifications (PTMs) of FVIII. Various host cell lines are available for recombinant protein expression for example bacterial, yeast, fungal, plant, insect and mammalian. Individual cell lines differ in yield, cost and stability, but most importantly, in their ability to perform PTMs. PTMs are the biochemical modifications in a protein after its translation from DNA to amino acid sequence. All human plasma proteins undergo PTMs. Different expressions systems (e.g. human or hamster cell lines) produce different PTMs for the same protein sequence. There are different types of PTMs, which include sulphation and glycosylation. Glycosylation alters the structural and functional properties of a protein and is responsible for physicochemical properties, immunological properties, receptor binding/affinity and intracellular sorting. For rfviii, the PTMs such as glycosylation and sulphation have been shown to be vital for functionality and VWF-binding affinity. In a study by Leyte et al. [1] it was shown that tyrosine sulphation at all six sites of the FVIII molecule is required for full FVIII activity. The absence of sulphation at Tyr 1680 reduces the affinity of FVIII for VWF fivefold. The other sites of tyrosine sulphation within FVIII affect the rate of cleavage by thrombin at the respective thrombin cleavage sites. It was concluded that sulphation of Tyr 1680 is required for the interaction of FVIII with VWF. Human-cl rhfviii is fully sulphated at Tyr 1680. A recent study by Kannicht et al. [2] to characterize PTMs of Human-cl rhfviii showed the product to be fully sulphated at the functionally critical sites and to be glycosylated with complex-, hybrid- and high mannose type glycosylation present at the same sites as in human plasma-derived FVIII. Sandberg et al. [3] measured the affinity of Human-cl rhfviii, ReFacto â, Advate â and Kogenate â to immobilized VWF by surface plasmon resonance. In additional experiments, CNBr Sepharose was covalently coupled with purified pdvwf, and rfviii products were added. After binding, residual FVIII:C in the supernatant was determined and plotted related to the applied FVIII:C. Human-cl rhfviii was shown to have a higher affinity to VWF than comparative rfviii products, thus minimizing circulating unbound FVIII and further reducing the potential risk of inhibitor development. Human-cl rhfviii was shown to be highly pure, with host cell protein and DNA traces comparable to, or lower than, currently marketed rfviii products. Human-cl rhfviii was shown to have high specific FVIII activity and characteristics similar to full-length rfviii products. The study by Kannicht et al. [2] showed N-glycan structures of the complex- and high-mannose type at the glycosylated asparagine residues Asn 41, Asn 239, Asn 1810 and Asn 2118 in Human-cl rhfviii as depicted in Fig. 1. Most importantly, rfviii expression in a human cell line avoids expression of the antigenic carbohydrate epitopes Gala1-3Galb1-GlcNAc-R (a-gal) and N-glycolylneuraminic acid (Neu5Gc) which are present on hamster glycoproteins, for example from baby hamster kidney or Chinese hamster ovary (CHO) cells, respectively (Fig. 2, [4,5]). These Fig. 1. Human-cl rhfviii N-glycosylation. Glycan structures at the glycosylated asparagine residues in Human-cl rhfviii. Glycans identified by LC-MS/MS at each of the four glycosylation sites. From [2] with permission. Haemophilia (2014), 20 (Suppl. 1), 1--9 2014 John Wiley & Sons Ltd

THE FIRST TRULY HUMAN RECOMBINANT FVIII 3 Fig. 2. N-glycosylation antigenic carbohydrate epitopes. antigenic epitopes are not present on Human-cl rhfviii. Anti-a-Gal is the most abundant natural antibody in all humans (~1% of circulating immunoglobulins in humans [6]). Anti-a-Gal mediates the rejection of pig xenograft organs in humans. The a-gal epitope has clinical potential in the production of vaccines expressing a-gal epitopes that can be targeted to antigen-presenting cells, thereby increasing the immunogenicity of viral and other microbial vaccines [7]. Different expression systems produce differently modified proteins from the same amino acid sequence. The high degree of sulphation at Tyr 1680 ensures high VWF-binding affinity and thus minimal levels of circulating unbound rhfviii. Both complete sulphation and the absence of antigenic carbohydrate epitopes aim to minimize the intrinsic immunogenicity of Human-cl rhfviii. Efficacy and safety of Human-cl rhfviii in adult patients: on demand vs. prophylaxis Prophylaxis with FVIII is considered the optimal treatment for managing patients with haemophilia A. Although there is ample evidence to support prophylactic treatment with FVIII in children with severe haemophilia A, adults with the disease are mainly treated on demand and the potential benefit of regular prophylaxis is linked to a higher consumption of costly FVIII concentrates. The completed Phase II (GENA-01) and III (GENA-08) studies evaluated the pharmacokinetics, safety, immunogenicity and efficacy of Human-cl rhfviii in the treatment and prophylaxis of bleeding in previously treated adult and adolescent patients with severe haemophilia A. GENA-01 was a prospective, randomized, actively controlled, open-label crossover multicenter ondemand trial which involved nine centres in three countries (USA, Germany and Bulgaria). It recruited 22 previously treated patients (PTPs) with severe haemophilia A (mean age 39.6 14.06). Its objectives were pharmacokinetic evaluation, efficacy of ondemand treatment and surgical prophylaxis, incremental recovery of FVIII:C, immunogenicity and safety. GENA-08 was a prospective, single arm, open-label, multinational prophylaxis trial involving 11 centres in four countries (Austria, Bulgaria, Germany and UK) which recruited 32 PTPs with severe haemophilia A (mean age 37.3 13.6). The primary objective was efficacy of prophylactic treatment, treatment of breakthrough bleeding and surgical prophylaxis. The secondary objectives were incremental recovery of FVIII:C, immunogenicity and safety. Both studies were similarly designed with regard to the inclusion/exclusion criteria, study duration and in how the data were captured and analyzed. In addition, the study populations were favourably comparable to each other with regard to age, body mass index, haemophilia joint health score (according to Hilliard et al. [8]), race and historical bleeding sites. In GENA-01, 65.1% of the bleeding episodes were spontaneous. The most common sites of bleeding were the knee (23.3%), elbow (22.8%) and ankle (15.7%). Of the bleeding, 57.4% was moderate to major and 42.2% was minor. There were 986 total bleeding episodes of which 94.4% were successfully treated (excellent or good response with excellent being abrupt pain relief and/or unequivocal improvement in objective signs of bleeding within approximately 8 h after a single infusion, and good being definite pain relief and/or 2014 John Wiley & Sons Ltd Haemophilia (2014), 20 (Suppl. 1), 1--9

4 L. A. VALENTINO et al. improvement in signs of bleeding within approximately 8 12 h after an infusion requiring 2 infusions for complete resolution). A total of 96.7% of all bleeding episodes were managed with one (90.3%) or two (6.4%) infusions. The median dose per infusion was 30 IU kg 1 (range 7 61) and the median dose per bleeding episode was 30.9 IU kg 1 (range 8 657). The total factor consumption was 156.9 IU kg 1 month 1. In GENA-08, there was good adherence with 93.6% of prophylactic infusions given every 2 days, for a median of 3.4 infusions per week (range 2.01 3.46). The dose per infusion was 33.1 IU kg 1 (median, range 24 39.3). There were 44 breakthrough bleeding events; most (59.1%) were spontaneous and only 28 required treatment. Of the 44 haemorrhages, 28 (63.6%) were minor. There was no major or lifethreatening bleeding and 50% of the patients (16/32) did not experience any bleeding episodes during the prophylactic treatment with Human-cl rhfviii. Before study entry patients treated on demand (n = 11) had a mean monthly bleeding rate of 3.92 episodes. This was reduced to 0.04 with Human-cl rhfviii. Patients on prophylaxis had a mean monthly bleeding rate of 0.54 episodes before study entry; this was reduced to 0.26 during prophylaxis with Human-cl rhfviii (Fig. 3). The efficacy of Human-cl rhfviii in the treatment of breakthrough bleeding was rated as excellent in 71.4% of cases and good in 28.6% of cases. 88.9% of all bleeding episodes were managed with one (81.5%) or two (7.4%) infusions. The median dose per infusion was 32.1 IU kg 1, range 20 53. In the attempt to compare GENA-01 and GENA-08 study results, adult patients on prophylactic regimens with Human-cl rhfviii experienced significantly fewer bleeding episodes than those using on-demand treatment (Fig. 4). There was a 96% reduction in mean monthly bleeding rates with prophylaxis vs. ondemand treatment (4.77 vs. 0.19, Fig. 4), whereas factor consumption increased by a factor of 3 (156.9 vs. 474.1 IU kg 1 month 1 ). In summary, these studies in adult and adolescent patients with severe haemophilia A indicated that Human-cl rhfviii is safe and effective in the prevention and treatment of bleeding episodes. There were no product-related serious adverse events and none of the PTPs treated with Human-cl rhfviii developed inhibitors or an allergic reaction. It is concluded that prophylaxis with Human-cl rhfviii, in comparison to on-demand treatment in a comparable cohort, appears to prevent >90% of bleeding episodes in adults with severe haemophilia A. Human-cl rhfviii in children: results from the Phase III study A completed Phase III study (GENA-03) investigated the pharmacokinetics, efficacy, safety and immunogenicity of Human-cl rhfviii in previously treated children aged 2 12 years with severe haemophilia A. The primary objectives of the trial were to assess the efficacy of prophylactic treatment and the treatment of Fig. 3. Reduction in annual bleeding rate under prophylaxis with Humancl rhfviii: historical vs. GENA-08 study data. Fig. 4. Mean monthly bleeding rate per regimen with Human-cl rhfviii: on demand vs. prophylaxis. Adult patients on prophylactic regimens experienced significantly fewer bleeding episodes than those using on-demand treatment. Haemophilia (2014), 20 (Suppl. 1), 1--9 2014 John Wiley & Sons Ltd

THE FIRST TRULY HUMAN RECOMBINANT FVIII 5 Fig. 5. Mean plasma FVIII:C level for Human-cl rhviii (standardized to 50 IU kg 1 ): chromogenic and one-stage assay (n = 26). breakthrough bleeding. The secondary objectives were to measure pharmacokinetics in different age groups, incremental recovery of FVIII:C, immunogenicity, efficacy during surgery, safety and tolerability. Fifty-nine patients from 15 sites in seven countries (UK, Poland, France, Russia, Turkey, Romania and Czech Republic) were enroled. Thirteen children from each age group (younger group I: 2 5 years, older group II: 6 12 years) participated in the comparative pharmacokinetic investigation. Prophylactic treatment was given as 30 40 IU kg 1 every other day or three times per week for 6 months and 50 exposure days. With regard to demographics the mean age of the children was 6.1 years (range 2 12); weight was 26.7 kg (range 8 73), all patients were white and 53 patients (89.8%) were on prophylaxis prior to the start of the study. Mean pharmacokinetic parameters of Human-cl rhfviii were similar for the chromogenic and the onestage assay (Fig. 5). In younger children (2 5 years old) the half-life was 9.49 3.32 h for the chromogenic assay and 11.91 5.36 for the one-stage assay. The corresponding figures in older children (6 12 years old) were 9.99 1.88 h and 13.08 2.59 h, respectively. In vivo recovery remained stable throughout the study. Compliance to the treatment schedule was very good (97%).* The mean average dose per infusion was 38.9 7.2 IU kg 1, and there was a mean of 3.1 0.3 infusions per week. The mean average amount per month was 527.7 112.3 IU kg 1. There were 108 treated bleeding episodes during the study. Of these 65 (60.2%) were traumatic and 36 (33.3%) were spontaneous. Of the haemorrhages, 61 (56.5%) were minor and 46 (42.6%) were moderate to major. There were no major to life-threatening haemorrhages. The most frequent sites of bleeding were the ankle (n = 21 bleeding episodes) and knee (n = 15 bleeding episodes). In the younger children the number of spontaneous and traumatic bleeding episodes was roughly equal (1.08 and 1.37 mean annual bleeding rate, respectively), but in the older children there were more traumatic bleeding episodes than spontaneous ones (3.26 and 1.90, respectively). Joint bleeding accounted for about 40% of all haemorrhages. In six children treated on demand before study entry there was a 97% reduction in the mean annual bleeding rate following treatment with Human-cl rhfviii from 35.9 episodes to 0.97 (P = 0.031). In 52 children on prophylaxis (data not available for one child) there was a reduction of 14% in the mean annual bleeding rate, from 5.11 episodes to 4.38 (non-significant). The results with Human-cl rhfviii compare very favourably with bleeding rates from other studies such as the Guardian TM 3 study [9] which investigated the effect of turoctocog alfa (CHO cell line) in children. The latter study looked at a comparable patient population of PTPs with severe haemophilia with regard to age, weight and race. It found that the annual bleeding rate in young children was 4.73, compared to 2.57 in GENA-03, and in older children the rate was 5.86 with turoctocog alfa compared to 5.58 with Human-cl rhfviii (Fig. 6 [9]). In GENA-03 no related serious adverse events were reported and none of the children developed anti- Human-cl rhfviii antibodies or inhibitors. Human-cl rhfviii was well tolerated with no cases of allergic reactions. Human-cl rhfviii was efficacious and safe in the prevention and treatment of bleeding episodes in previously treated children. The results support further investigation of Human-cl rhfviii in PUPs. Due to the absence of immunogenic epitopes seen in rfviii concentrates from hamster cell * A total of 5316 infusions for prophylaxis were recorded in the prophylaxis population; only 172 infusions (3.2%) were administered more than 3 days after the previous infusion (3 days is the maximum possible gap in a three times weekly schedule). 2014 John Wiley & Sons Ltd Haemophilia (2014), 20 (Suppl. 1), 1--9

6 L. A. VALENTINO et al. Fig. 6. Annual bleeding rates from GENA-03 and Guardian TM 3. lines, Human-cl rhfviii is thought to be potentially less immunogenic. A Phase III prospective, multicentre, multinational, open-label, non-controlled clinical trial in PUPs to evaluate immunogenicity, safety, tolerability and efficacy of Human-cl rhfviii in the prophylaxis, treatment of bleeding episodes and in surgical prophylaxis has been initiated (GENA-05, NuProtect). Can new research techniques enable future strategies to avoid inhibitor formation? Understanding the immunological mechanism of inhibitor formation remains challenging. The risk of developing an inhibitor depends on multiple genetic (F8 mutation, F8 haplotype, ethnicity, human leucocyte antigen [HAL] class II and immune response genes) and non-genetic factors (infection/inflammation and peak factor levels). The long-term goal for haemophilia treaters would be to have prospective predictors of inhibitor formation or avoidance; prospective predictors of immune tolerance induction (ITI), success or failure, and to identify interventions for high-risk patients. This section briefly summarizes the background rationale for the satellite RNA sequencing (RNA seq) scientific substudy in the forthcoming NuProtect trial and how this technology may contribute to future understanding of inhibitor formation and tolerization mechanisms. There has been rapid progress in the field of genomics in the last 60 years. In 1986, the human genome project began and in 2001 the initial human genome sequence was published [10], 2 years earlier than predicted and less than 50 years since the basic structure of DNA was described [11]. Further rapid technological advance, most noteably that of next-generation sequencing (NGS), makes high throughput sequencing accessible and affordable. NGS enables genome-wide, RNA seq offering the chance to apply this exciting technology in key studies. RNA seq captures dynamic gene expression, that is which genes are actually being used or suppressed at the point of sampling. A genome-wide assessment of gene use (transcriptome analysis) will provide a snapshot of the environment of FVIII concentrate exposure in PUPs, combining genetic data reflecting both environmental factors (inflammation, infection) and inherited genetic polymorphisms. RNA seq will be carried out in the NuProtect trial as a satellite scientific study. It will be the first study in PUPs utilizing this technology. Such a study will create a large and valuable data set for future research. It will be hypothesis generating, given the unselected, genome-wide recording of RNA expression data. In PUPs, 1 ml of additional blood for sequencing will be taken at baseline and every 3 4 exposure days with routine inhibitor screening until 20 exposure days. RNA studies have been used in vaccine research to understand the mechanisms by which vaccines stimulate protective immunity. Data have enabled prediction of the immunogenicity or efficacy of vaccines and the technique has already defined predictive signatures of human antibody responses to influenza vaccination. A study by Bucasas et al. [12] investigated the correlation of gene expression patterns and antibody response to influenza vaccination in a cohort of healthy male adults. The study showed that markedup regulation of expression of genes involved in Haemophilia (2014), 20 (Suppl. 1), 1--9 2014 John Wiley & Sons Ltd

THE FIRST TRULY HUMAN RECOMBINANT FVIII 7 interferon signalling, positive IL-6 regulation and antigen processing and presentation were detected early, within 24 h of vaccination. Later RNA signatures involved cellular proliferation, protein metabolism and antiapoptosis pathways. The authors concluded that high vaccine responder status correlates with increased early expression of interferon signalling and antigen processing and presentation genes. The NuProtect RNA seq satellite scientific substudy hypothesizes that RNA seq can identify similar pathways that will influence and possibly predict inhibitor formation in the PUP setting, and tolerization in the ITI setting. Blood volumes required for RNA seq are small and applicable to the paediatric setting. The entire RNA transcriptome will be analyzed in the context of first FVIII exposure. This will generate the first genome-wide RNA seq data sets in persons with haemophilia, including gene expression markers for both adaptive and innate immunity. The data obtained from this study and accompanying scientific satellite studies (FVIII genotyping, FVIII haplotyping and in vitro T-cell assays) should provide key insights into the genetic dynamics of inhibitor formation and tolerization. The vision for these satellite studies is for translation into patient benefits by identifying those at risk of inhibitor formation and ultimately contributing to a reduction in the frequency of inhibitors. Octapharma s commitment to such a study, generating vast data sets of gene usage at key treatment time points, will create a very important resource for future research. Bringing Human-cl rhfviii to PUPs and personalized prophylaxis The finalized clinical development programme with Human-cl rhfviii for registration in the USA and Europe was discussed with the FDA and took into consideration the current Committee for Medicinal Products for Human use (CHMP) European Medicines Agency (EMA) guidelines for FVIII concentrates. Endpoints and assays were harmonized so that data between studies can be compared. Clinical studies with Human-cl rhfviii began in 2009 with GENA-09 which took place in Russia involving 22 PTPs aged 18 years. It was a crossover pharmacokinetic study investigating prophylaxis, breakthrough bleeding and surgery. Patients were treated for 50 exposure days and for 6 months and the study was completed in 2010. Two further studies followed: GENA-01, a pharmacokinetic investigational new drug multinational study involving 22 PTPs aged 12 65 years; this was a crossover pharmacokinetic study that began in 2010 with patients treated on demand and during surgery for 50 exposure days and treatment lasting for 6 months. The study was completed at the end of 2012. The other study that began in 2010 was GENA-08, an EU study involving 32 PTPs aged 18 75 years given prophylaxis with Human-cl rhfviii, and treatment for breakthrough bleeding and surgery for 50 exposure days lasting 6 months. This study was also completed in 2012. GENA-03, one of the first paediatric studies conducted according to the new CHMP guideline began in 2011 and was completed in 2013, this was an EU study involving 59 paediatric PTPs aged 2 to <12 years. It looked at pharmacokinetics vs. previous FVIII, using prophylaxis and treatment for breakthrough bleeds and during surgery. Again the study was for 50 exposure days and for 6 months. GENA-13, the continuation study of GENA-03, began in 2013 and will be ongoing until launch. GENA-05 (NuProtect) The PUP study NuProtect also began in 2013. It was a requirement to initiate a PUP study before EMA submission in Europe. Based on the absence of nonhuman immunogenic epitopes as seen in other rfviii concentrates from hamster cell lines, it is hypothesized that Human-cl rhfviii may be less immunogenic. This hypothesis will be tested in the Phase III prospective, multicentre, multinational PUP study NuProtect, which will involve 16 countries (including Brazil, Canada, France, Germany, India, the UK and USA) and 45 centres worldwide. It is planned to enrol 100 haemophilia A PUPs who will be under observation for their first 100 exposure days or a maximum study participation of 5 years. The trial will look at immunogenicity, efficacy (during prophylaxis, treatment of breakthrough bleeding and during surgery), safety and tolerability of Human-cl rhfviii and will run until Q4 2018. The trial will also look at health economic modelling analysis with resource use parameters. With regard to immunogenicity NuProtect will measure inhibitors using the Nijmegen modification and anti-fviii antibodies using an ELISA-based screening method. Measurements will be taken at screening then every 3 4 exposure days until exposure Day 20, then every 10 12 exposure days until exposure Day 100 and then every 3 months until study completion. Gene mutation analysis will also be made during the study. Optional investigations/substudies are as follows: 1. Recovery investigation. 2. Immunogenotyping (to investigate genetic factors that might influence the development of FVIII inhibitors, using HLA typing, immune response genes, and F8 ethnic haplotypes). 3. In vitro immunogenicity (to assess the nature of T-cell response by analysing cytokine expression [interlukin (IL)-2, TNFa, IFNc, IL-5, IL-6, IL-10 and IL-17] and T-cell proliferation). 2014 John Wiley & Sons Ltd Haemophilia (2014), 20 (Suppl. 1), 1--9

8 L. A. VALENTINO et al. 4. Epitope mapping (to investigate the antibody response/specificity against FVIII). 5. RNA expression profiling (to provide an understanding of the transcript activity of the genes involved in immune responses that may be responsible for FVIII inhibitor formation). Inclusion criteria are male patients with severe haemophilia A (FVIII:C < 1%) who have not previously received treatment with FVIII concentrates or other blood products containing FVIII. Fully informed written and signed consent obtained before the study commences is mandatory. Treatment options are shown in Table 1. By end of June 2013, 11 PUPs were enroled and 4 of these have started treatment. Seventeen centres in seven countries have been initiated. GENA-21(NuPreviq) In addition to the PUP study and based on the long half-life of Human-cl rhfviii of a mean of 17.1 h (range, 11 24 h, median 13.7 h, IQR 11.97 17.50 from the GENA-01 pharmacokinetic study, see Fig. 7), an individualized prophylaxis study was initiated in 2013, NuPreviq (GENA-21). NuPreviq is a prospective, open-label, multicentre Phase IIIb study to assess the efficacy and safety of individually tailored prophylaxis with Human-cl Table 1. Treatment options in NuProtect (GENA-05). rhfviii in adult PTPs with severe haemophilia A. The goal of this study is to provide optimal treatment for each patient based on their own pharmacokinetic properties by maximizing the treatment interval to prevent bleeding. The primary objective is to compare the annual bleeding rate of individualized tailored prophylaxis treatment with the historical annual bleeding rate from the on-demand study GENA-01 (58.1 haemorrhages per year). Secondary objectives are to compare the spontaneous annual bleeding rate of individualized tailored prophylaxis with the historical annual bleeding rate from the on-demand study GENA-01 (38.5 haemorrhages per year); to compare the annual bleeding rate in patients on twice weekly or less prophylaxis with the historical annual bleed rate from the GENA-01 study and to assess the pharmacokinetics of Human-cl rhfviii. Individualized prophylaxis will be given for 6 months and the observation period will be 8 months on average for each patient. Further individualized prospective measures have been integrated into the study protocol such as thrombin generation assay (TGA) evaluation to analyze the correlation among FVIII plasma levels, thrombin generation potential and the frequency of breakthrough bleeding events. The trial will recruit 50 evaluable adult PTPs and will run until 2015. Twentyseven centres in 10 countries (including the UK, Spain, Germany and Poland) have been recruited. Recommended dose Start Frequency of treatment Frequency or dose adjustments Prophylactic treatment >20 IU FVIII kg 1 BW With the first BE Depends on the patient s clinical situation At investigator s discretion Minor haemorrhage Moderate to major haemorrhage Major to life-threatening haemorrhage On-demand treatment Recommended dose 20 30 IU FVIII kg 1 BW 30 40 IU FVIII kg 1 BW 50 80 IU FVIII kg 1 BW Frequency Every 8 24 h until BE is resolved Every 6 24 h until BE is resolved Repeat >20 IU FVIII kg 1 BW every 6 12 h until BE is resolved BE, bleeding episode; BW, body weight. Fig. 7. Elimination half-lives of Human-cl rhfviii in GENA-01 according to the one-stage clotting assay. Haemophilia (2014), 20 (Suppl. 1), 1--9 2014 John Wiley & Sons Ltd

THE FIRST TRULY HUMAN RECOMBINANT FVIII 9 In the initial pharmacokinetic evaluation phase, Human-cl rhfviii will be given for 72 h at 60 5IUkg 1. In Phase I of the prophylactic treatment, Human-cl rhfviii will be administered at 30 40 IU kg 1 every other day or three times a week for 1 3 months until the individual pharmacokinetic variables are analyzed. Once available, patients will switch to Phase II, the individually tailored prophylaxis treatment that will be given for 6 months. Trough and peak FVIII:C and TGA will be measured at 2, 4 and 6 months. Both NuProtect and NuPreviq have the potential to show the favourable features of the first recombinant and unmodified FVIII from a human cell line. Conclusion The available data show that this human rfviii product is not associated with the development of inhibitors or serious adverse reactions in 135 PTPs treated so far. Its efficacy is clearly stated and prophylaxis is associated with very high efficacy rates in various haemophilia populations. Immunogenicity and pharmacokinetics have to be further investigated, and it will be extremely interesting to see if there is a link between the production of a recombinant human FVIII in a human cell line and immunogenicity. In this context, PTMs, which are differently introduced on the recombinant protein by human cell lines and murine cell lines, might play a major role. NGS will be useful in the future to better guide the use of factor concentrates and better identify the genes and proteins implicated in protection or predisposition with regard to inhibitor formation. Additional prospective studies with Human-cl rhfviii are needed to gain further insight in the treatment of haemophilia, but the results from clinical trials accumulated so far look extremely promising. References 1 Leyte A, van Schijndel HB, Niehrs C et al. Sulfation of Tyr1680 of human blood coagulation factor VIII is essential for the interaction of factor VIII with von Willebrand factor. J Biol Chem 1991; 266: 740 6. 2 Kannicht C, Ramstr om M, Kohla G et al. Characterisation of the post-translational modifications of a novel, human cell linederived recombinant human factor VIII. Thromb Res 2013; 131: 78 88. 3 Sandberg H, Kannicht C, Stenlund P et al. Functional characteristics of the novel, human-derived recombinant FVIII protein product, human-cl rhfviii. Thromb Res 2012; 130: 80817. 4 Hironaka T, Furukawa K, Esmon PC et al. Comparative study of the sugar chains of Acknowledgements Writing support was provided by Ros Kenn, freelance medical editor/ writer and funded by Octapharma. Disclosures factor VIII purified from human plasma and from the culture media of recombinant baby hamster kidney cells. J Biol Chem 1992; 267: 8012 20. 5 Shen BW, Spiegel PC, Chang CH et al. The tertiary structure and domain organization of coagulation factor VIII. Blood 2008; 111: 1240 7. 6 Galili U. The alpha-gal epitope and the anti-gal antibody in xenotransplantation and in cancer immunotherapy. Immunol Cell Biol 2005; 83: 674 86. 7 Macher BA, Galili U. The Galalpha1, 3Galbeta1,4GlcNAc-R (alpha-gal) epitope: a carbohydrate of unique evolution and clinical relevance. Biochim Biophys Acta 2008; 1780: 75 88. 8 Hilliard P, Funk S, Zourikian N et al. Hemophilia joint health score reliability study. Haemophilia 2006; 12: 518 25. Leonard A. Valentino has received fees for consulting, and scientific advisory from Baxter Healthcare Corporation, Bayer HealthCare Pharmaceuticals, Biogen Idec, CSL Behring, GTC Biotherapeutics, Inspiration Biopharmaceuticals, Novo Nordisk and Pfizer; fees for presentations from Pfizer and training funds from Baxter Healthcare Corporation. Claude Negrier has received research support from Alnylam, Baxter, Bayer, Biogen Idec, CSL Behring, Inspiration, LFB, Novo Nordisk, Octapharma and Pfizer; travel support from CSL Behring, Novo Nordisk, SOBI/Biogen Idec; consultancy fees from Alnylam, Baxter, Bayer, Biogen Idec, CSL Behring, Inspiration, LFB, Novo Nordisk, Pfizer; honoraria from Baxter, Bayer, Biogen Idec, CSL Behring, Inspiration, LFB, Novo Nordisk, Octapharma and Pfizer and is on scientific advisory boards for Alnylam, Baxter, Bayer, Biogen Idec, CSL Behring, Inspiration, LFB, Novo Nordisk and Pfizer. Guido Kohla works for Octapharma AG. He has no other conflicts of interest to declare. Andreas Tiede has received research support from Baxter, Bayer, CSL Behring, Novo Nordisk, Octapharma and Pfizer; travel support from Baxter, Bayer, CSL Behring, Novo Nordisk, Octapharma and Pfizer; consultancy fees from Baxter, Bayer, CSL Behring, Novo Nordisk, Octapharma and Pfizer and honoraria from Baxter, Bayer, CSL Behring, Novo Nordisk, Octapharma and Pfizer. Raina Liesner has received research support for clinical trials for Biogen, Inspiration, Octapharma and Pfizer; travel support from Bayer, Novo Nordisk and Octapharma; honoraria from Baxter, Bayer, Novo Nordisk and Octapharma and is on scientific advisory boards for Baxter, Bayer, Novo Nordisk and Pfizer. Dan Hart has received research support from Bayer (Early Career Investigator) and Octapharma; travel support from Octapharma and honoraria from Baxter, Bayer, Novo Nordisk and Pfizer. Sigurd Knaub works for Octapharma AG. He has no other conflicts of interest to declare. Declaration of funding interests: full funding was provided by Octapharma. 9 Kulkarni R, Karim FA, Glamocanin S et al. Results from a large multinational clinical trial (guardian TM 3) using prophylactic treatment with turoctocog alfa in paediatric patients with severe haemophilia A: safety, efficacy and pharmacokinetics. Haemophilia 2013; 19: 698 705. 10 Lander ES, Linton LM, Birren B et al.; International Human Genome Sequencing Consortium. Initial sequencing and analysis of the human genome. Nature, 2001; 409: 860 921. 11 Watson JD, Crick FHC. A structure for deoxyribose nucleic acid. Nature 1953; 171: 737 8. 12 Bucasas KL, Franco LM, Shaw CA et al. Early patterns of gene expression correlate with the humoral immune response to influenza vaccination in humans. J Infect Dis 2011; 203: 9219. 2014 John Wiley & Sons Ltd Haemophilia (2014), 20 (Suppl. 1), 1--9