RECOMBINANT ALBUMIN FUSION PROTEINS - Fusion Protein Technologies for Biopharmaceuticals: Applications and Challenges - page 166

Nevertheless, 10-30% of patients develop inhibitory anti-
bodies against exogenous FVIII [72]. For these patients, the
use of recombinant coagulation factors with the ability to
bypass FVIII in the coagulation cascade, such as activated
recombinant factor VII (rFVIIa, NovoSeven 1 ), can effec-
tively restore homeostasis [73]. The half-life of rFVIIa is

rFVIIa [11]. The results suggest that a single injection of
rVIIa-FP may be sufficient per bleeding episode.

10.3.2.10 Factor IX Hemophilia B is a congenital coag-
ulation disorder characterized by a lack or deficiency of
coagulation factor IX (FIX). The management of hemophilia
B includes treatment with plasma-derived FIX (e.g., Mono-
nine 1 ) or recombinant FIX (rFIX; BeneFIX). For rFIX, the
half-life is 18-34 h [77,78] and therefore, when applied in a
prophylactic regimen, requires intravenous administration
every 2-3 days to prevent spontaneous bleeding [79]. The
need for a long-acting preparation of rFIX that would allow
for less frequent dosing prompted the development of
recombinant FIX albumin fusion proteins [12,80].
Novel recombinant FIX albumin fusion proteins using
linker sequences that contain the same cleavage site as that
responsible for proteolytical activation of FIX were devel-
oped [12]. Activation of FIX by either FVIIa/tissue factor
(TF) or FXIa would therefore also cleave the linker, sepa-
rating the FIX and rHA moieties of the fusion protein. Thus,
FIX would be cleaved from albumin when activated, allow-
ing for enhanced biological activity compared to previous
FIX fusion proteins [12]. Because of the complex post-
translational modifications of the FIX moiety, the fusion
proteins were expressed in mammalian cell lines [12].
Recombinant FIX albumin fusion proteins with cleavable
linkers had 10- to 30-fold higher molar specific activity in a
one-stage clotting assay compared to those with noncleav-
able linkers [12]. Fusion proteins with cleavable linkers also
showed greater effects in a thrombin generation assay than
did those with noncleavable linkers. Despite the improve-
ment in biological activity compared to fusion proteins with
noncleavable linkers, the molar specific activity of fusion
proteins with cleavable linkers still remained lower than that
of rFIX or pdFIX in vitro. This suggests that albumin can
interfere with the activation process of FIX to some extent,
possibly because of steric hindrance [12,81].
Similar to rVIIa-FP, rIX-FP demonstrated improved phar-
macokinetics in rats, rabbits, and FIX / mice [12,82].
Specifically, the half-life, recovery, and AUC of rIX-FP
were superior to those of rFIX (BeneFIX) (Table 10.2) [12].
In a tail-tip bleeding model using FIX-deficient mice,
rIX-FP with cleavable linker reduced bleeding time in a
dose-dependent manner that was comparable to that of rFIX
[12]. The dose calculations were based on a FIX one-stage
clotting assay. At a dose of 100 IU per kg body weight, the
bleeding time was reduced almost to that of normal mice,
which had a mean bleeding time of around 2min. Moreover,
the cleavable rIX-FP minimizes the risk of generating excess
FIXa fusion proteins with extended half-life, thereby reduc-
ing the potential risk of ensuing prothrombotic effects [12].

2.5 h, and frequent injections are therefore necessary and
long-term prophylaxis is hardly feasible [74].
A novel rFVIIa fusion protein (rVIIa-FP) was developed
using a flexible glycine/serine linker fused between the C-
terminus of rFVIIa and recombinant albumin [11]. The
length of the linker (31 amino acids) was chosen to minimize
potential interactions between the two proteins and to opti-
mize the activity of FVIIa [10,11,76]. The construct was
produced in mammalian cell lines (HEK-293 and CHO). A
eukaryotic expression system was required to achieve the
functionally important posttranslational modifications of the
fusion protein, including g -carboxylation, N- and O-glyco-
sylation, and b -hydroxylation [10].
The pharmacokinetics of rVIIa-FP was assessed in rats
[11,39], mice [39], and rabbits [39,75]. In all animal species,
the fusion protein demonstrated significant improvements in
pharmacokinetic parameters (Table 10.2).
In an in vitro evaluation of rVIIa-FP in human FVIII
inhibitor whole blood, rVIIa-FP produced a dose-dependent
reduction in clot formation time that was similar to that of
the wild-type protein [11]. The activity of rVIIa-FP was also
evaluated in vivo using rats pretreated with phenprocoumon,
an anticoagulant agent that blocks the synthesis of FVII over
a period of 16 h. When administered 15.75 h after phenpro-
coumon administration, rVIIa-FP corrected whole blood
clotting time measured by thrombelastography to a similar
degree as did rFVIIa, confirming the hemostatic activity of
the fusion protein. In addition, when rVIIa-FP or rFVIIa was
administered immediately after phenprocoumon, the fusion
protein effectively corrected clotting time 16 h later, but
rFVIIa did not, because of its short half-life. Thus, the fusion
protein was shown to be efficacious in an in vivo situation
and its effects endured much longer than did those of

TABLE 10.2 Pharmacokinetic Parameters of Activated
Recombinant Factor VII Albumin Fusion Protein (rVIIa-FP)
and Recombinant Factor IX Albumin Fusion Protein (rIX-FP)
Compared with NovoSeven 1 and BeneFIX 1 , Respectively, in
Different Animal Species [11,12,39]

Ratio Fusion Protein/
Wild-Type Protein

rVIIa-FP rIX-FP

Species Mouse Rat Rabbit Rat Rabbit

AUC 11.4 14.5 14.2 4.6 7.2

Terminal half-life 4.1 5.8 8.1 4.7 4.0

In vivo recovery 2.0 2.4 2.0 1.7 1.6

10.3.2.11 Infestin-4 Infestin is a Kazal-type serine pro-
tease inhibitor found in the midgut of the hematophagus

Data are expressed as ratios of rVIIa-FP to NovoSeven and rIX-FP to
BeneFIX. AUC, area under the curve.