Journal of Thrombosis and Haemostasis, 6: 277–283
ORIGINAL ARTICLE
Safety and pharmacokinetics of a recombinant factor VIII with pegylated liposomes in severe hemophilia A J . S . P O W E L L , * D . J . N U G E N T , J . A . H A R R I S O N , * A . S O N I , A . L U K , à H . S T A S S § and E . G O R I N A à *Division of Hematology and Oncology, University of California at Davis, Sacramento, CA; Department of Hematology, ChildrenÕs Hospital of Orange County, Orange, CA; àBayer HealthCare Pharmaceuticals, Hematology/Cardiology, Berkeley, CA, USA; and §Bayer HealthCare AG, Wuppertal, Germany
To cite this article: Powell JS, Nugent DJ, Harrison JA, Soni A, Luk A, Stass H, Gorina E. Safety and pharmacokinetics of a recombinant factor VIII with pegylated liposomes in severe hemophilia A. J Thromb Haemost 2008; 6: 277–83.
Introduction Summary. Background: BAY 79-4980 is a sucrose-formulated recombinant factor VIII (rFVIII-FS) combined with pegylated liposomes to prolong activity. Objectives: To investigate the safety, tolerability, bioavailability, pharmacokinetics and pharmacodynamics of a single administration of BAY 79-4980 compared with standard rFVIII-FS in patients with severe hemophilia A. Methods: This randomized, double-blind study consisted of two crossover substudies comparing two doses of liposomal rFVIII-FS with standard rFVIII-FS. Males (12– 60 years) with severe hemophilia A received a single infusion of standard rFVIII-FS (35 IU kg)1) followed by a single infusion of BAY 79-4980 (13 or 22 mg kg)1 pegylated liposomes) or vice versa, with 12 observation days and a 2-day washout period between treatments. Results: Twenty-six subjects were enrolled at two centers. No serious adverse events were reported. Transient increases in complement C3a, but not CH50, were seen in subjects receiving both the low- and high-liposome-dose BAY 79-4980. Mild transient elevations of total and lowdensity lipoprotein cholesterol were observed. There were no clinically significant differences in clotting or laboratory parameters or in pharmacokinetic behavior between BAY 794980 and standard rFVIII-FS. The number of subjects with spontaneous bleeds on days 1–14 postinfusion was low, and group comparisons were inconclusive. Conclusions: Singledose administration of BAY 79-4980 is well tolerated in patients with severe hemophilia A. Plasma pharmacokinetics of FVIII cannot explain the extended protection from bleeding observed previously with BAY 79-4980. Further studies of efficacy and long-term safety of chronic administration are planned. Keywords: hemophilia, liposome, longer-acting factor VIII, pegylation, pharmacokinetics, safety.
Correspondence: Jerry Powell, UC Davis Hemophilia Treatment Center, 2360 Stockton Blvd, Suite 1100, Sacramento, CA 95817, USA. Tel.: +1 916 734 8616; fax: +1 916 734 7946; e-mail:
[email protected] Received 12 June 2007, accepted 31 October 2007 2007 International Society on Thrombosis and Haemostasis
Patients with severe hemophilia A are at high risk for spontaneous hemorrhaging into soft tissues and joints that can culminate in debilitating arthropathy [1,2]. Prevention of spontaneous bleeding episodes in patients with severe hemophilia A usually requires therapeutic infusions of coagulation factor VIII (FVIII) every 2–3 days due to the short half-life of FVIII in circulation (12 h) [3–5]. Although effective in reducing complications related to bleeding episodes, the frequent i.v. infusions that such prophylaxis regimens require place a considerable burden on patients and their caregivers and may reduce compliance with therapy [6,7]. The development of new FVIII products with prolonged activity would allow less frequent infusions and could increase the effectiveness of hemophilia A prophylaxis through improved compliance. Current technologies that extend the activity of other therapeutic compounds include liposomal formulation and the incorporation of high-molecular-weight polyethylene glycol (PEG). Liposomes are small (100-nm diameter) vesicles that consist of a phospholipid bilayer surrounding an aqueous interior. Liposomes have been effective as delivery vehicles for numerous drugs, particularly those used in cancer therapy [8], but are rapidly cleared from circulation [9]. High-molecularweight PEG polymers have also been used to extend the circulatory availability of therapeutic molecules by either direct conjugation to the active molecule or through conjugation to the liposome carrier [10]. BAY 79-4980 (Bayer HealthCare, Berkeley, CA, USA) is a recombinant, sucrose-formulated FVIII product (rFVIII-FS; Kogenate FS) combined at reconstitution with a pegylated liposome carrier. The rFVIII-FS molecules bind non-covalently to PEG on the liposome surface with high affinity (Kd 1.9–4.5 nM) and without disrupting either the binding capacity of FVIII to its circulatory carrier protein, von Willebrand factor (VWF), or coagulation activity [11]. In vivo efficacy was evaluated in a hemophilic mouse model that demonstrated modestly increased circulatory half-life of BAY 79-4980 compared with rFVIII-FS and significantly better survival of BAY 79-4980-infused mice compared with rFVIII-FS-infused
278 J. S. Powell et al
mice [11]. The magnitude and timing of this protective effect did not correlate with the modest increase in plasma half-life. An early clinical evaluation also showed significantly improved efficacy of prophylactic infusion of BAY 79-4980 over rFVIII-FS [12]. In this controlled crossover study, 24 adult subjects with severe hemophilia A were infused with BAY 794980 or rFVIII-FS, and the time to their next spontaneous bleed was measured. Compared with subjects treated with rFVIII-FS, BAY 79-4980-treated subjects had nearly double the mean length of time free from bleeding episodes. Statistically significant, dose-dependent improvements were reported at the two dose levels tested (25 IU kg)1 and 35 IU kg)1). Furthermore, no clinically significant adverse events were reported, suggesting that two repeated infusions of BAY 794980 were well tolerated. The current study investigated the safety, tolerability, bioavailability, pharmacokinetics and pharmacodynamics of a single administration of BAY 79-4980 compared with rFVIII-FS. Methods Study participants
The study enrolled previously treated males with ‡ 200 exposure days (EDs) to replacement FVIII products, including 20 EDs in the previous 12 months, aged 12–60 years with severe hemophilia A (< 1% baseline plasma FVIII). Subjects who had no history of FVIII-inhibitor antibody formation and were FVIII-inhibitor antibody-negative (< 0.6 BU mL–1) at screening, no signs or symptoms of an acute bleeding episode on the day of infusion, and no FVIII treatment within 4 days before infusion were eligible. Subjects were excluded if they: had acquired immunodeficiency syndrome (AIDS), abnormal renal function, active hepatic disease, anemia, thrombocytopenia, hematologic/bleeding conditions other than hemophilia A, high blood pressure, or dyslipidemia; had a history of severe reaction(s) to or requirement of premedication for FVIII products; were unable to go without FVIII treatment for at least 4 days prior to study entry or between study infusions; had been treated with interferon or other experimental drugs within the previous 3 months; or had a known allergy or severe reactions to liposomes or PEG. Study design
This phase I randomized, double-blind, crossover study was conducted at two centers in the USA in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. The study protocol was approved by appropriate institutional review boards and independent ethics committees at the participating centers, and all subjects gave written informed consent. Following screening, subjects were assigned to one of four treatment sequences according to a predefined, site-specific randomization schedule (Fig. 1). All subjects received rFVIII-
Screening Day –1 to –7
Period A
Day 1
Washout Day 12
rFVIII-FS (n = 6)
Period B
Day 1
High-concentration BAY 79–4980
High-concentration BAY 79–4980 (n = 7)
rFVIII-FS (n = 6)
Day 22 to 24
rFVIII-FS Low-concentration BAY 79–4980
Low-concentration BAY 79–4980 (n = 7)
Randomization
Follow-up Day 12
rFVIII-FS*
Crossover
* One patient withdrew from the study following period A
Fig. 1. Study design.
FS infusions at 35 IU kg)1, reconstituted in 2.5 mL sterile water for injection per 1000 IU, and either high-dose [35 IU FVIII kg–1 in 22 mg liposomes kg–1 (2450 IU in a 17.1 mL volume for a 70-kg person)] or low-dose [35 IU FVIII kg–1 in 13 mg liposomes kg–1 (2450 IU in a 9.8 mL volume for a 70-kg person)] BAY 79-4980. These doses were expected to increase the plasma concentration of FVIII to a level of about 60–80% of normal physiological FVIII activity. The study drugs were administered by slow i.v. infusion of the first 5 mL at 22 mL h)1, and the next 5 mL at 44 mL h)1, for a total infusion time of 30 min. Subjects and all site personnel were blinded to the study drug, except the pharmacist who reconstituted the medication and the infusionist who administered the medication. Because the appearance and reconstituted volume of the study drugs differed, the infusion was performed from behind a screen to maintain the blind. In the event of a bleeding episode during period A or the washout period, the subjectÕs prestudy FVIII preparation was administered. No further FVIII pharmacokinetic data were collected; however, liposome pharmacokinetic data were collected until the end of the period. Subjects underwent a washout interval of ‡ 4 days from any FVIII treatment before entering period B. Safety parameters
The main safety endpoints were the frequency, severity and duration of treatment-related adverse events during and after a single i.v. infusion of low-liposome-dose BAY 79-4980, highliposome-dose BAY 79-4980, or standard rFVIII-FS. Safety was further assessed by monitoring bleeding episodes, hematology and clinical chemistry, lipid panel, coagulation parameters, complement activation, FVIII inhibitor levels (Nijmegenmodified Bethesda assay), urinalysis, vital signs and electrocardiogram (ECG). Pharmacokinetic and pharmacodynamic parameters
The primary pharmacokinetic (PK) objective was to compare the PK profiles of the low- and high-dose BAY 79-4980 with that of standard rFVIII-FS; a secondary objective was to 2007 International Society on Thrombosis and Haemostasis
Safety and pharmacokinetics of PEG-Lip rFVIII 279
characterize the liposomal PK profiles. The following parameters were calculated using non-compartmental analytical methods using KinCalc (Bayer Healthcare, Leverkusen, Germany) from measurements using the chromogenic as well as the one-stage clotting assays: incremental recovery (10 min postinfusion); area under the curve from time 0 to 24 h (AUC0– 24), from time 0 to last measurable concentration (AUC0–tn), and from time 0 to infinity (AUC0–¥); maximum plasma concentration (Cmax); terminal elimination half-life (t1/2); mean residence time (MRT); clearance (CL); and volume of distribution at steady-state (Vss). Plasma levels of the liposome components palmitoyl oleoyl phosphatidyl choline (POPC) and MPEG-2000-DSPE [N-(carbonyl-methoxypolyethyleneglycol-2000)-1,2-distearoyl-sn-glycero-3-phosphoethanolamine] served as surrogate markers for the liposomal PK analyses. The following liposome PK parameters were calculated: AUC0–tn, AUC0–¥, Cmax, t1/2, MRT, CL and Vss. Where appropriate, PK estimates based on baseline-corrected concentrations were performed to yield technically correct parameter evaluations. Blood samples were obtained pre-infusion and at 10 min, 30 min and 1, 3, 6, 12, 24, 48, 72 (day 3), 96 (day 4), 120 (day 5) and 168 h (day 7) postinfusion for FVIII PK (one-stage clotting and chromogenic assays) and liposome PK (POPC and MPEG-2000-DSPE measurements). Additional blood samples were obtained at 240 h (day 10) and 288 h (day 12) postinfusion for liposome PK. Statistical analysis
All statistical computations were performed using SAS version 9.0 (SAS Institute Inc, Cary, NC, USA). The safety analysis included all subjects randomized into the study who received any amount of the study drug. For the pharmacokinetic evaluations, AUC and Cmax values were subjected to an ANOVA according to the study design. Ninety per cent confidence intervals were calculated for the ratio of the geometric means between low-liposome-dose BAY 79-4980 and standard
rFVIII-FS, and between high-liposome-dose BAY 79-4980 and standard rFVIII-FS. Results Subject characteristics
Twenty-six subjects were randomized to treatment and included in the safety analysis (Fig. 1, Table 1). One subject in treatment sequence 4 withdrew from the study during period A because of an adverse event and was included only in the safety analysis. Another subject (treatment sequence 1) was excluded from the pharmacokinetic analysis because he had pre-infusion FVIII levels of 5–7% despite a documented prestudy diagnosis of severe hemophilia A. Baseline characteristics were similar between sequence groups (Table 1). Safety analysis
With the exception of the subject who withdrew from the study, all other subjects received both scheduled infusions. For the high-liposome-dose group there were 13 exposures each to BAY 79-4980 and to rFVIII-FS. In the low-liposome-dose arms, there were a total of 13 exposures to BAY 79-4980 and 12 exposures to rFVIII-FS. Seven of the 26 subjects experienced a total of eight adverse events (Table 2), five of which were considered by the investigators to be possibly drug related. No adverse event was considered severe or serious. One subject discontinued the study during period A (low-liposome-dose BAY 79-4980 infusion) because of increased breathing frequency and flushing that were possibly drug related. After delivery of about 75% of the medication, the infusion was terminated and not reinitiated, and the symptoms (nausea, moderate difficulty breathing, feeling flushed, reddened sclerae, and prominent flushing on the upper torso, neck and head) resolved without medical intervention.
Table 1 Baseline demographics and subject characteristics Treatment group
Mean age, years ± SD Mean body weight, kg ± SD Race, n (%) White Black Asian Hispanic HCV-ab positive HIV-ab positive
rFVIII-FS fi low-liposome-dose BAY 79-4980 (n = 6)
Low-liposome-dose BAY 79-4980 fi rFVIII-FS (n = 7)
rFVIII-FS fi high-liposome-dose BAY 79-4980 (n = 6)
High-liposome-dose BAY 79-4980 fi rFVIII-FS (n = 7)
27.8 ± 10.1 84.4 ± 26.7
22.3 ± 4.9 81.6 ± 20.1
22.7 ± 3.1 76.4 ± 20.5
19.1 ± 3.3 86.4 ± 26.2
2 0 1 3 4 0
(33) (17) (50) (67)
6 (86) 1 (14) 0 0 5 (71) 0
HCV-ab, hepatitis C virus antibody; HIV-ab, human immunodeficiency virus antibody. 2007 International Society on Thrombosis and Haemostasis
4 1 0 1 4 1
(67) (16) (17) (67) (17)
5 (71) 0 0 2 (29) 3 (43) 0
280 J. S. Powell et al Table 2 Treatment-emergent adverse events Number of subjects, n (%)
Type of adverse event
rFVIII-FS Low-liposome-dose rFVIII-FS High-liposome-dose (low-liposome-dose comparator) BAY 79-4980 (high-liposome-dose comparator) BAY 79-4980 (n = 12) (n = 13) (n = 13) (n = 13)
Any adverse event Otitis externa Limb injury Increased blood triglycerides Headache Pharyngolaryngeal pain Respiratory distress
1 (8) 0 0 0 0 1 (8) 0
3 1 0 0 1 0 1
(23) (8)
(8)* (8)*
2 (15) 0 1 (8) 1 (8)* 0 0 0
2 (15) 0 0 2 (15)* 0 0 0
*Possibly drug related.
Mean total cholesterol levels increased consistently in both BAY 79-4980 liposome-dose groups in a liposome-dosedependent manner and reached their maximum levels at 6 h (low dose) and 24 h (high dose) postinfusion (Fig. 2). There was a median increase of 15 and 20 mg dL)1 over baseline in the low- and high-liposome-dose arms, respectively. Mean cholesterol levels after 120 h remained slightly elevated compared with mean pre-infusion values but returned to baseline by the follow-up visit. Mean peak cholesterol levels for the total population remained below 200 mg dL)1, though the peak levels for nine subjects exceeded 200 mg dL)1. The increase in total cholesterol levels was due predominantly to increased low-density-lipoprotein (LDL) cholesterol. No effects on mean high-density-lipoprotein (HDL) cholesterol levels were observed. No increases in total cholesterol over baseline were seen after the rFVIII-FS infusions. Mean C3a levels increased in a dose-dependent manner within 10 min of BAY 79-4980 infusion and decreased to normal after 12 h. Transient C3a elevations in 9 of 13 (69%) subjects receiving BAY 79-4980 were considered relevant (at least 2-fold of the pre-infusion level and ‡ 65 ng mL)1). C3a increases were not associated with an increase in CH50 complement activity. The subject who discontinued treatment
Mean value (mg dL–1) ± SD
250
BAY 79–4980 (high) BAY 79–4980 (low)
rFVIII-FS (high) rFVIII-FS (low)
200
150
100
Screening Pre6h infusion
24 h
48 h
72 h
96 h
120 h Follow-up
Time after infusion
Fig. 2. Mean total cholesterol concentration over time. Dotted line represents the upper limit for desirable or optimal levels for persons with or without existing heart disease.
(see above) had a transient increase in C3a level from 31 ng mL)1 pre-infusion to 764 ng mL)1 at 10 min postinfusion. Because eight other subjects demonstrated transiently elevated C3a levels that did not coincide with the occurrence of an adverse event, this measure did not appear to be prognostic. No increases in mean C3a levels were observed in subjects receiving rFVIII-FS. No differences in coagulation parameters were observed between BAY 79-4980 and rFVIII-FS infusions. No subject developed FVIII inhibitors during the study. There were no clinically significant changes in D-dimer values or VWF antigen. Urinalysis did not indicate any relevant changes in specific gravity, pH, appearance, ketones, protein, blood, glucose, red blood cells, white blood cells, crystals, casts or bacteria. No clinically relevant changes in vital signs, hematology laboratory parameters and clinical chemistry (including liver function tests), ECG or physical examination findings were reported, except in the subject noted above. Pharmacokinetic parameters
In all groups, mean FVIII levels increased to a maximum within approximately 10 min after infusion and then decreased log-linearly over time (Fig. 3). The findings from the one-stage coagulation and chromogenic assays were comparable. No statistically significant differences in FVIII pharmacokinetics were observed between the low-liposome-dose BAY 79-4980 group and the rFVIII-FS group. For the high-liposome-dose BAY 79-4980 group, no statistically significant differences were observed after adjustment for multiple comparisons (Bonferroni criterion) (Table 3). With regard to the liposome PK characteristics, the presence of endogenous phospholipids did not allow resolution of POPC PK parameters. Because MPEG is a liposome component not normally found in plasma, it was the more meaningful parameter by which to evaluate the liposome PK. Mean AUC values were higher after administration of high-liposome-dose than low-liposome-dose BAY 79-4980. The observed half-life of MPEG was between 31 and 36 h, with an MRT of 47–52 h (Table 4). 2007 International Society on Thrombosis and Haemostasis
72
96
120
0.1
0
24
46
72
96
120
Times (hours) after infusion
Fig. 3. Mean plasma factor VIII activity over time using the one-stage coagulation assay (A) and the chromogenic assay (B). Geometric mean ± SD is based on 12 subjects per time point unless otherwise indicated.
Discussion This phase I crossover study compared the pharmacokinetic profile of FVIII after a single infusion of low- and high-dose BAY 79-4980, a pegylated liposome preparation incorporating rFVIII-FS, and standard rFVIII-FS in patients with severe hemophilia A. This study also assessed the safety of BAY 794980, with special regard to the lipid profile and adverse events. BAY 79-4980 infusions were generally well tolerated. No serious adverse events were reported. Three of five adverse events considered possibly related to the study drug were triglyceride elevations. Because therapy with BAY 79-4980 would involve the regular infusion of lipids into the circulation, changes in blood lipids were a focus of the safety evaluation. Fluctuation in mean triglyceride levels was observed in the trial but did not correlate with infusion administration or with a specific preparation. The variation in mean triglycerides was likely to be related to diet. A dose-dependent increase in mean cholesterol levels of approximately 8% and 20% from baseline occurred within the first 24 h postinfusion with BAY 79-4980 at low- and high-liposome doses, respectively, and returned to baseline thereafter. The increase was attributable mainly to a temporary rise in LDL cholesterol. A transient increase in circulating cholesterol is known to be associated with liposome administration and may represent redistribution of cholesterol from the tissues to the liposomes, as liposomes do not inherently contain cholesterol [13]. Mean peak cholesterol levels, however, remained below 200 mg dL)1, which is considered a reasonable cutoff for desirable or optimal 2007 International Society on Thrombosis and Haemostasis
rFVIII-FS, geometric mean (approx % CV) [range] BAY 79-4980, geometric mean (approx % CV) [range]
n=8 n=9 n = 10 n=8
1
LS mean ratio [90% CI] (P-value)à
n=9 n = 11 n = 11 n = 10
rFVIII-FS, geometric mean (approx. % CV) [range]
n = 11 n = 11 n = 12 n = 11
BAY 79-4980, geometric mean (approx % CV) [range]
n = 11 n = 11 n = 12 n = 11
10
Low-dose groups (n = 12)
100
High-dose groups (n = 12)*
rFVIII-FS (high) rFVIII-FS (low)
BAY 79–4980 (high) BAY 79–4980 (low)
Table 3 Pharmacokinetic parameters for FVIII using the one-stage coagulation assay. Between-group differences were comparable using the chromogenic assay
Mean FVIII activity (0.01 IU dL–1)
B 1000
Based on the results of measured FVIII levels
46
Times (hours) after infusion
§
24
ANOVA.
0
AUC, area under the curve; CL, clearance; MRT, mean residence time; t½, half-life; Vss, volume of distribution at steady state. *AUC0–24 in the BAY 79-4980 group was evaluated in 11 subjects. AUC0–24 in both groups was evaluated in 11 subjects. àMeasured using (not corrected for baseline).
0.1
n=8 n=9 n = 10 n=8
(0.065) 1126.9 (26.2) [727.6–1803.0] 1143.5 (22.8) [883.4–1730.0] 1.01 [0.96–1.08] (0.663) (0.107) 875.2 (20.9) [627.6–1124.0] 881.6 (18.2) [693.9–1146.0] 1.01 [0.96–1.07] (0.767) (0.042) 1153.4 (27.2) [761.0–1864.0] 1135.5 (28.6) [625.8–1794.0] 0.98 [0.87–1.11] (0.882) (0.070) 246.9 (25.9) [174.2–368.5] 242.0 (24.8) [160.2–388.1] 0.98 [0.93–1.04] (0.524) (0.070) 3.13 (25.8) [1.94–4.81] 3.07 (23.5) [2.00–4.08] 0.98 [0.93–1.04] (0.525) (0.836) 89.0 (19.3) [71.7–132.6] 87.0 (17.4) [66.4–122.0] 0.98 [0.90–1.06] (0.623) (0.707) 16.0 (22.5) [10.9–24.7] 16.1 (21.5) [10.2–22.3] 1.01 [0.96–1.05] (0.753) (0.098) 50.0 (19.0) [32.5–64.2] 49.4 (19.4) [32–67.1] 0.99 [0.95–1.03] (0.564) (0.302) 11.4 (21.8) [8.0–17.1] 11.6 (22.0) [7.2–16.0] 0.98 [0.93–1.04] (0.587)
n=9 n = 11 n = 11 n = 10
1
[1.01–1.10] [1.00–1.10] [1.02–1.16] [0.91–0.99] [0.91–0.99] [0.90–1.14] [0.96–1.03] [0.90–1.00] [0.88–1.03]
n = 11 n = 11 n = 12 n = 11
1.05 1.05 1.09 0.95 0.98 1.01 0.99 0.95 0.95
n = 11 n = 11 n = 12 n = 11
10
[496.0–1482.0] 991.2 (31.2) [537.3–1454.0] [451.3–1072.0] 821.8 (24.1) [491.6–1075.0] [542.5–1540.0] 1057.1 (29.9) [602.3–1586.0] [198.8–441.7] 281.2 (27.4) [166.4–415.1] [2.32–6.96] 3.53 (31.7) [2.37–6.54] [63.4–120.9] 91.3 (21.0) [66.1–145.3] [8.5–19.5] 13.3 (22.9) [9.3–19.0] [39.8–73.0] 47.0 (13.6) [40.1–63.2] [6.4–21.1] 9.6 (22.4) [6.5–13.6]
100
(31.8) (25.7) (31.6) (26.1) (31.7) (19.7) (24.8) (16.4) (31.0)
rFVIII-FS (high) rFVIII-FS (low)
954.3 779.7 989.2 291.8 3.66 92.1 13.4 49.2 10.2
BAY 79–4980 (high) BAY 79–4980 (low)
AUC0–¥ (0.01 IU*h mL–1) AUC0–24§ (0.01 IU*h mL–1) AUC0–tn§ (0.01 IU*h mL–1) CL (mL h)1) CLnorm (mL h–1 kg–1) Cmax (0.01 IU mL)1) MRT (h) Vss (mL kg)1) t1/2 (h)
A 1000
LS mean ratio [90% CI] (P-value)à
Mean FVIII activity (0.01 IU dL–1)
Safety and pharmacokinetics of PEG-Lip rFVIII 281
282 J. S. Powell et al Table 4 Pharmacokinetic parameters for MPEG
AUC0–¥ (mcg*h mL–1) AUC0–24* (mcg*h mL–1) AUC0–tn* (mcg*h mL–1) AUCtn–¥ (%) Cmax (lg mL)1) MRT (h) t1/2 (h)
High-liposome-dose BAY 79-4980, geometric mean (approx. % CV) [range] (n = 12)*
Low-liposome-dose BAY 79-4980, geometric mean (approx. % CV) [range] (n = 12)
3207 1233 2483 17.9 79.4 46.6 31.0
1742 600 927 38.3 36.9 52.4 35.9
(59.3) (40.0) (70.6) (52.2) (39.1) (44.3) (46.1)
[1126–7457] [620–2056] [850–4287] [7.2–37.8] [47.4–178.9] [15.3–96.5] [10.5–62.7]
(48.3) (33.9) (72.8) (47.4) (35.4) (41.7) (43.0)
[735–3173] [358–887] [352–2325] [13.0–65.2] [20.3–55.5] [29.5–118.5] [20.1–83.9]
AUC, area under the curve; Cmax, maximum concentration; t½, half-life; MRT, mean residence time. *AUC0–24 and AUC0–tn were evaluated in seven subjects. AUCtn–¥ was evaluated in 12 subjects.
long-term levels for persons with or without existing heart disease [14]. Potential for immunogenicity was evaluated by formation of inhibitors to FVIII. No subject developed an FVIII inhibitor during this study. Complement activation was followed as a possible indicator of hypersensitivity, because complement can be activated via both the classic and alternative pathways, giving rise to C3a and C5a anaphylatoxins that may trigger hypersensitivity reactions. A transient dose-dependent increase in mean C3a levels peaked at 10 min postinfusion and returned to baseline within 12 h. Elevated C3a levels did not correlate with adverse events. C3a elevation has been observed following infusion of FDA-approved liposomal drugs in the past, without any detectable clinical problems [15]. This elevation might represent an acute-phase reaction or part of early, inconsequential complement activation. One subject in this study developed transient increased breathing with nausea and moderate flushing soon after BAY 79-4980 was initiated. His C3a levels increased in parallel, suggesting a hypersensitivity reaction to BAY 79-4980. Complement activation and associated reactions, called complement activation-related pseudoallergy (CARPA), have been previously described in association with the administration of pegylated liposomal drugs [15]. Contrary to a classic type I hypersensitivity reaction, CARPA is not IgE-mediated, occurs at the first antigen exposure, and is mild or absent on repeated exposure. The reaction observed in this subject is consistent with CARPA, but the subject was not re-exposed. Because only 1 of 26 subjects in this study experienced such a reaction, and increases in C3a were observed in 69% of subjects (some higher than this subjectÕs peak levels), increases in C3a might not be a specific diagnostic marker for this type of hypersensitivity reaction. The clinical efficacy of BAY 79-4980 was not an endpoint of this study. However, a controlled crossover clinical study designed to compare the efficacy of BAY 79-4980 with that of rFVIII-FS by measuring the time to next spontaneous bleeding episode following a single prophylactic infusion was recently published [12]. In the efficacy study, the postinfusion bleed-free period was significantly longer for subjects treated with BAY 79-4980 than for subjects treated with rFVIII-FS. Based on this
finding, the authors suggested that BAY 79-4980 might confer the benefits of prophylaxis with once-weekly or less frequent dosing. Our results suggest that BAY 79-4980 has a safety and tolerability profile that may be appropriate for prophylaxis, although larger clinical trials are needed to establish the safety and efficacy of repeated administration of BAY 79-4980. Analysis of FVIII PK parameters in plasma did not show any relevant differences between the BAY 79-4980 groups and their respective rFVIII-FS controls. Notably, the addition of pegylated liposomes did not appear to affect the hemostatic properties of rFVIII-FS. These findings suggest that any possible increased protection from bleeding obtained from BAY 79-4980 is not due to its PK properties. Additional investigations into the mechanism of action of BAY 79-4980 are therefore necessary. PK data for the MPEG component of the liposome showed a delayed distribution profile as initially expected for a 2000 kD PEG-free molecule, suggesting a marked influence of the bound liposomes and/or a redistribution effect as a result of other interactions. The finding that rFVIII-FS and MPEG demonstrate different PK parameters (e.g. time–concentration curves) after BAY 79-4980 infusion suggests that the FVIII and MPEG components are metabolized by independent pathways in the circulation after infusion. The influence of liposomes on FVIII is not apparent from a PK point of view, and the actual mechanism of action for BAY 79-4980 remains to be elucidated. Possibly, rFVIII-FS with pegylated liposomes localizes to specific cells such as platelets, which in turn create local, small concentrations of rFVIII-FS upon platelet activation and aggregation at sites of injury. Hemostasis is a complicated summation of multiple biochemical reactions, as evidenced in part by the great variability in clinical bleeding among patients with hemophilia. Other coagulation interactions clearly are involved to spare some of these patients severe bleeding complications. In conclusion, this study demonstrates that single administration of BAY 79-4980 appears to be well tolerated, and its plasma PK profile is similar to that of conventional rFVIII-FS. If the safety and efficacy profile of BAY 79-4980 is confirmed by larger trials that involve longer-term administration, then BAY 79-4980 may be suitable for prophylactic FVIII treatment 2007 International Society on Thrombosis and Haemostasis
Safety and pharmacokinetics of PEG-Lip rFVIII 283
with longer intervals between infusions. Development of a prophylactic therapy that requires less frequent dosing than current FVIII preparations could have a notable effect on treatment compliance and, hence, on clinical outcomes and burden of illness in patients with severe hemophilia A. Acknowledgements The authors thank A. Lau and B. Castro for editorial support on this manuscript. Disclosure of Conflict of Interests Funding for this study was provided by Bayer HealthCare Pharmaceuticals, Hematology/Cardiology. J. S. Powell and D. J. Nugent are consultants for Bayer HealthCare. A. Luk, H. Stass and E. Gorina are employees of Bayer HealthCare. References 1 Roosendaal G, Mauser-Bunschoten EP, de Kleijn P, Heijnen L, van den Berg HM, van Rinsum AC, Lafeber FP, Bijlsma JW. Synovium in haemophilic arthropathy. Haemophilia 1998; 4: 502–5. 2 Fischer K, Bom JG, Mauser-Bunschoten EP, Roosendaal G, Berg HM. Effects of haemophilic arthropathy on health-related quality of life and socio-economic parameters. Haemophilia 2005; 11: 43–8. 3 Berntorp E, Astermark J, Bjorkman S, Blanchette VS, Fischer K, Giangrande PL, Gringeri A, Ljung RC, Manco-Johnson MJ, Morfini M, Kilcoyne RF, Petrini P, Rodriguez-Merchan EC, Schramm W, Shapiro A, van den Berg HM, Hart C. Consensus perspectives on prophylactic therapy for haemophilia: summary statement. Haemophilia 2003; 9 (Suppl. 1): 1–4.
2007 International Society on Thrombosis and Haemostasis
4 Bjorkman S. Prophylactic dosing of factor VIII and factor IX from a clinical pharmacokinetic perspective. Haemophilia 2003; 9 (Suppl. 1): 101–8. 5 Bjorkman S, Berntorp E. Pharmacokinetics of coagulation factors: clinical relevance for patients with haemophilia. Clin Pharmacokinet 2001; 40: 815–32. 6 Hacker MR, Geraghty S, Manco-Johnson M. Barriers to compliance with prophylaxis therapy in haemophilia. Haemophilia 2001; 7: 392–6. 7 Claxton AJ, Cramer J, Pierce C. A systematic review of the associations between dose regimens and medication compliance. Clin Ther 2001; 23: 1296–310. 8 Park JW, Benz CC, Martin FJ. Future directions of liposome- and immunoliposome-based cancer therapeutics. Semin Oncol 2004; 31 (Suppl. 13): 196–205. 9 Senior JH. Fate and behavior of liposomes in vivo: a review of controlling factors. Crit Rev Ther Drug Carrier Syst 1987; 3: 123–93. 10 Bhadra D, Bhadra S, Jain P, Jain NK. Pegnology: a review of PEGylated systems. Pharmazie 2002; 57: 5–29. 11 Baru M, Carmel-Goren L, Barenholz Y, Dayan I, Ostropolets S, Slepoy I, Gvirtzer N, Fukson V, Spira J. Factor VIII efficient and specific non-covalent binding to PEGylated liposomes enables prolongation of its circulation time and haemostatic efficacy. Thromb Haemost 2005; 93: 1061–8. 12 Spira J, Plyushch OP, Andreeva TA, Andreev Y. Prolonged bleedingfree period following prophylactic infusion of recombinant factor VIII (Kogenate FS) reconstituted with pegylated liposomes. Blood 2006; 108: 3668–73. 13 Carpentier YA, Richelle M, Bihain BE, Dahlan W, Haumont D, Deckelbaum RJ. Interactions between exogenous fat and plasma/ lipoproteins. Infusionsther Klin Ernahr 1987; 14 (Suppl. 3): 29–32. 14 Centers for Disease Control and Prevention. High Blood Cholesterol Prevention. http://www.cdc.gov/cholesterol/prevention.htm; accessed 8 December 2006. 15 Szebeni J. Complement activation-related pseudoallergy: a new class of drug-induced acute immune toxicity. Toxicology 2005; 216: 106–21.
