Characteristics of Patients without Bleeding in a Pivotal Trial of Extended Half-Life, Pegylated, Full-length Recombinant Factor VIII (BAX 855) in the Treatment of Hemophilia A

BAX 855 is a pegylated full-length recombinant factor VIII (PEG rFVIII) built on ADVATE (rFVIII) with an extended half-life and is in development for prophylaxis and treatment of bleeding in patients with hemophilia A. In the pivotal trial, previously treated patients (PTPs) received BAX 855 prophylaxis with a twice-weekly 45 IU/kg dosing regimen. Of 101 patients in the per protocol analysis set, 40 (39.6%) patients achieved zero bleeding during 6 months of prophylactic treatment with BAX 855, compared with the on-demand treatment arm in which every patient (n=17) experienced at least 1 bleeding episode.

Pharmacokinetic parameters were assessed in a subset of patients with zero bleeding (n=9). Assessments were performed for rFVIII, for an initial infusion of BAX 855 and for a repeat infusion of BAX 855 after 6 months of treatment. Using the one-stage clotting assay, BAX 855 demonstrated an extended circulation time compared to rFVIII with a mean (standard deviation [SD]) half-life (T_1/2) of 16.13 (3.827) hours vs 11.44 (2.250) hours (ratio BAX 855/rFVIII: 1.4). Over time, the 6 month assessment of BAX 855 T_1/2 remained stable at 16.30 (3.137) hours (ratio repeat/initial BAX 855: 1.0), based on the one-stage clotting assay. Incremental recovery (IR) measurements were comparable with 2.41 (0.454) IU/dL for rFVIII and 2.62 (0.684) IU/dL for BAX 855. As expected, clearance (CL) was higher for rFVIII compared to BAX 855: 0.0380 (0.009) vs 0.0205 (0.007) dL/(kg*h).

Historical annualized bleeding rates (ABRs) in the 40 patients without bleeding during prophylactic treatment were calculated based on the previous 3-6 months of each patient’s diary or recall. Historical ABRs ranged from 0-80 with a median ABR of 7. However, patients may have either been receiving on demand or prophylactic FVIII treatment prior to entering the prophylactic arm of the trial, accounting for the range in ABRs. The median pre-trial ABR of patients in the on-demand arm, all of whom experienced bleeding, was much higher at 41.5 (range: 12.9-67.9). Of the 40 patients without bleeding who were in the prophylaxis arm of the trial, 31 received prophylaxis in their previous pre-trial regimen, while 9 had received on-demand treatment prior to enrolling in the trial. 

A target joint was defined as a single joint (ankles, knee, hip, or elbow) with ≥ 3 spontaneous bleeding episodes in any consecutive 6-month period. At screening, 30% of patients without bleeding on the study had 0 target joints, while 37.5% had 1-2 and 32.5% had ≥3 target joints. Of the patients who had >3 target joints at screening, there were 4 (10%) patients with 4 and 1 (2.5%) patient with 6 target joints at screening. In contrast, patients in the on-demand group (who had all been treated on-demand prior to the trial) had a higher incidence of target joints at screening: 11.8% of patients had 0 target joints, while 52.9% had 1-2 and 28.3% had ≥3 target joints.

The presence or absence of preexisting arthropathy did not seem to influence the likelihood of having zero bleeding during the trial. In 65% of the patients without bleeding, arthropathy was present at screening. This is higher than was found in the patients treated on-demand (47.1%) who all experienced bleeding. A patient was considered to have arthropathy if it was indicated in the medical history or if the patient had joint surgery.

Blood group types were collected on the patients without bleeding. The results showed: A-50%; B-12.5%; AB-7.5%; O-15%; unknown 15%. This is rather different than the blood types of the 17 on demand patients who all experienced joint bleeding: A-35.3%; B-5.9%; AB-17.6%; O-35.3%; unknown 5.9%. There were fewer blood group O patients among the patients without bleeding.

While some of the characteristics of patients on prophylaxis without bleeding episodes were similar to those of patients treated on-demand  who all experienced bleeding, patients without bleeding had fewer target joints at screening, lower median historical ABR, and lower incidence of blood type O.