Long‐term patterns of safety and efficacy of bleeding prophylaxis with turoctocog alfa (NovoEight®) in previously treated patients with severe haemophilia A: interim results of the guardian™2 extension trial

M. OZELO,* M. MISGAV,† F. ABDUL KARIM,‡ S. R. LENTZ,§ M. MARTIN-SALCES,¶ I . MATYTSINA,** T. SAUGSTRUP** and E. SANTAGOSTINO†† *IHTC ‘Claudio L.P. Correa’, INCT do Sangue Hemocentro UNICAMP, University of Campinas, Sao Paulo, Brazil; †National Hemophilia Center, Chaim Sheba Medical Center, Tel Aviv University, Tel Aviv, Israel; ‡National Blood Centre, Kuala Lumpur, Malaysia; §The University of Iowa Carver College of Medicine, Iowa City, IA, USA; ¶Haematology Department, Hospital Universitario La Paz, Madrid, Spain; **Novo Nordisk A/S, Soeborg, Denmark; and ††Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, Maggiore Hospital Policlinico, Milan, Italy

Haemophilia A (factor VIII deficiency) is the most common type of haemophilia [1], affecting approximately 1 in 5000 males, although its prevalence varies globally [2]. Restoration of haemostasis with FVIII concentrates is the standard therapy. Turoctocog alfa (NovoEight â ; Novo Nordisk A/S, Bagsvaerd, Denmark) is a third generation human recombinant B domain-truncated FVIII molecule [3] that in Phase 3 trialsguardian TM 1 (adolescent and adult patients ≥12 years) [4] and guardian TM 3 (paediatric patients <12 years) [5] demonstrated favourable efficacy and safety profiles; there were no inhibitors and no safety concerns were observed. Subjects who completed the guardian TM 1 or guardian TM 3 trials could choose to continue treatment with turoctocog alfa by enrolling in the open-label, non-controlled guardian TM 2 extension trial (www.clinicaltrials.gov: NCT00984126) that investigates the long-term safety and efficacy of turoctocog alfa administered as prophylaxis and for treatment of bleeds.
This letter presents a preplanned interim analysis of data collected up to September 1, 2012 for the guardian TM 2 extension trial. The primary objective is to assess the safety of turoctocog alfa for prevention and treatment of bleeds. The primary endpoint is the frequency of developing FVIII inhibitors. Key efficacy endpoints are haemostatic effect during treatment of bleeding, number of infusions needed to treat a bleeding episode and annualized bleeding rate (ABR). Patients received turoctocog alfa as prophylactic treatment (20-50 IU kg À1 every second day or 20-60 IU kg À1 three times weekly) and for treatment of bleeds. Turoctocog alfa was in most cases administered at home by the patient or a caregiver. FVIII inhibitors were analysed at a central laboratory (Laboratorium f€ ur Klinische Forschung GmbH, Schwentinental, Germany) using the Nijmegen modification of the Bethesda assay [6,7]. In accordance with EMA guidance [8], a patient was considered to have a FVIII inhibitor if he tested positive [≥0.6 Bethesda units (BU)] in two consecutive samples.
Joint bleeds were categorized as target joint bleeds (target joint defined as ≥3 bleeding episodes in the same joint within a 6-month period prior to the trial) or non-target joint bleeds. The haemostatic effect of turoctocog alfa was self-assessed by the patient using a predefined 4-point scale (excellent, good, moderate and none; definitions are provided in the footnote of Table 1). The ratings 'excellent' and 'good' were considered treatment successes; ratings of 'moderate' and 'none' were considered treatment failures.
Evaluation of data was based on descriptive statistics. Data were evaluated for four predefined patient age groups (≤5 years; 6-11 years; 12-17 years; ≥18 years). ABR was estimated by cause of bleed ('spontaneous'; 'traumatic') and by location ('joint'; 'target joint'; 'non-joint') using a Poisson model allowing for overdispersion and presented with 95% confidence intervals; median values were also calculated for ABR.
As of 1 September 2012, the guardian TM 2 full analysis set comprised 188 patients, 1-61 years of age, with severe haemophilia A (133 adults and adolescents; 55 children). As of the cut-off date, 18 patients had been withdrawn during guardian TM 2 for: withdrawn consent (4 patients); treatment with another factor (4 patients); adverse event (AE) (3 patients); non-compliance (2 patients); transfer to another trial (2 patients); treatment with an anticoagulant agent (1 patient); unplanned surgery (1 patient); treatment for hepatitis C (1 patient).
This interim analysis includes approximately 3 years of data for the first patient in guardian TM 2, and a total of 255.9 patient-years exposure. The mean number of exposure days (EDs) was 205. Safety was assessed in all 188 patients exposed to turoctocog alfa; none developed FVIII inhibitors. Turoctocog alfa was considered to be well tolerated, with no unexpected patterns seen in AEs or serious adverse events (SAEs).
Seven AEs (peripheral oedema, increased aspartate aminotransferase, increased alanine aminotransferase, n, number of patients. Mucocutaneous was defined as a subcutaneous and/or mucosal bleed. A target joint was defined as a joint with three or more bleeds within 6 months. *Mild/moderate = minor bleeds which were uncomplicated joint bleeds, muscular bleeds without compartment syndrome, mucosal or subcutaneous bleeds. Severe = major bleeds which require hospitalisation, all head and neck bleeds, muscle bleeds with compartment syndrome, bleeds associated with a significant decrease in haemoglobin level (>3 g dL À1 ). All mild/moderate bleeds which were active after 24 h changed category to severe. † Classification of the haemostatic response of turoctocog alfa when used for treatment of bleeding episodes: excellent = abrupt pain relief and/or unequivocal improvement in objective signs of bleeding episode within approximately 8 h after a single infusion; good = definite pain relief and/or improvement in signs of bleeding episode within approximately 8 h after infusion, but possibly requiring more than one infusion for complete resolution; moderate = probable or slight beneficial effect within approximately 8 h after the first infusion, usually requiring more than one infusion; none = no improvement or worsening of symptoms. § Success defined as 'excellent' or 'good' haemostatic response. musculoskeletal pain, pain in extremity and two cases of arthropathy) in 4/188 (2.1%) patients were considered by the investigator to be 'possibly' or 'probably' related to turoctocog alfa. These events were nonserious and of mild or moderate severity. Three patients were withdrawn due to AEs: one for fatal trauma caused by alleged assault involving subdural haemorrhage, one for paranoid-type schizophrenia (patient had history of psychiatric disorders), and one for increased hepatic enzymes that developed at the end of the guardian TM 1 trial and precluded participation in the guardian TM 2 trial. No thromboembolic events or hypersensitivity reactions against turoctocog alfa were reported and no other safety concerns were identified. The mean per-patient consumption of turoctocog alfa for prophylaxis was 404 IU kg À1 per month, with a mean dose level of 31. 5  . Interestingly, these ABRs are lower than the ABRs reported in the guardian TM 1 and guardian TM 3 trials [4,5]. It may be that prolonging prophylaxis during the extension trial may have progressively decreased ABR. Additional differences were observed when ABR was classified by type of bleed among the age groups. Traumatic bleeding rates were highest among patients aged 6-11 years and lowest in adults, with the opposite pattern observed for spontaneous bleeding rates (Fig. 1a). Joint bleeding rates were higher than non-joint bleeding rates in all ages except children aged ≤5 years, for whom non-joint ABR was higher (rates were not compared statistically). Target joint bleeds comprised slightly more than half of joint ABR totals, across age groups, except 0-5 years (Fig. 1b).
A total of 752 bleeding episodes occurred in 142 (76%) patients during guardian TM 2; (Table 1). The majority (88%) of bleeding episodes were classified as mild or moderate. Twenty-four percent (24%) of patients did not experience a bleeding episode. However, some patients had a limited duration of exposure in the extension trial (exposure ranged from 1-492 days).
In all age groups, the majority of bleeds were joint bleeds (78%, overall); of these, 61% were in target joints. No non-traumatic intracranial haemorrhages occurred. Treatment was considered to be successful for 664/752 (88%) of overall bleeding episodes with minimal variation among age groups (Table 1). A total of 678 (90%) episodes were resolved with 1-2 infusions of turoctocog alfa. The overall mean consumption of turoctocog alfa to treat a bleed was 57.5 IU kg À1 body weight per episode, with slight variation observed among age groups.
In conclusion, this interim analysis demonstrated that the extended use of turoctocog alfa is safe and effective in prevention and treatment of bleeding episodes in adult, adolescent and paediatric patients. Turoctocog alfa was well tolerated and no patients developed FVIII inhibitors. In accordance with previously published studies, a prophylactic regimen with turoctocog alfa was shown to be beneficial in the treatment of severe haemophilia A.