Pharmacokinetics and safety of fibrinogen concentrate

Authors


Marilyn J. Manco-Johnson, Mountain States Regional Hemophilia & Thrombosis Center, Fitzsimons Building 500, Room WG109, 13001 East 17th Place, Aurora, CO 80010, USA.
Tel.: +1 001 303 724 0365; fax: +1 001 303 724 0947.
E-mail: Marilyn.Manco-Johnson@ucdenver.edu

Abstract

Summary. Background: Although fibrinogen concentrate has been available for the treatment of congenital fibrinogen deficiency for years, knowledge of its pharmacokinetics comes from only two small studies. Objectives: To assess the pharmacokinetic (PK) profile, clot integrity and safety of fibrinogen concentrate (human) (FCH) in patients with afibrinogenemia. Patients and methods: A multinational, prospective, open-label, uncontrolled study of patients with afibrinogenemia ≥ 6 years of age was conducted in the USA and Italy. Plasma was collected before and after infusion for PK analyses and evaluation by rotational thromboelastometry of maximum clot firmness (MCF) to assess clot integrity. Safety was assessed on the basis of adverse events and laboratory parameters. Results: After a single dose of 70 mg kg−1 body weight (b.w.) FCH in 14 patients, median incremental in vivo recovery was a 1.7 mg dL−1 increase per mg kg−1 b.w., and median levels were 1.3 g L−1 for fibrinogen activity and antigen 1 h after infusion. Median half-life (t1/2) was 77.1 h for fibrinogen activity and 88.0 h for antigen. Plasma recovery in children < 16 years old was similar to that in adults aged 16 to < 65 years, but the t1/2 and area under the curve were decreased, with an increased steady-state volume and clearance. MCF increased by a mean of 8.9 mm from baseline to 1 h after infusion of FCH (P < 0.0001). All four adverse events reported were mild, and none was serious or related to study drug. Conclusions: These PK findings confirm a rapid increase in plasma fibrinogen levels after infusion with FCH. Together with the clot integrity and safety data and published data on efficacy, the results support the idea that FCH substitution can restore hemostasis with a good safety profile.

Introduction

Normal coagulation may be disrupted by acquired or congenital fibrinogen deficiency, the latter comprising congenital afibrinogenemia, hypofibrinogenemia, and dysfibrinogenemia.

Currently, there are three therapeutic options for the treatment and prevention of bleeding in patients with congenital fibrinogen deficiency: plasma, cryoprecipitate, or fibrinogen concentrate. The therapeutic target for all of these options is to achieve a fibrinogen activity level of at least 1–1.5 g L−1 [1]. As compared with fibrinogen concentrates, infusion of plasma or cryoprecipitate is limited by lack of a virus inactivation/elimination process, leaving a risk of virus transmission, infusion of large amounts of unnecessary plasma proteins (all plasma proteins in plasma, and fibronectin, von Willebrand factor, factor VIII, factor XIII, and α-macroglobulins in cryoprecipitate) and anaphylatoxins, possibly causing allergic reactions, and the need to thaw plasma and cryoprecipitate before use and to administer them only in hospitals. As fibrinogen concentrate is markedly more concentrated (20 mg of fibrinogen mL−1 after reconstitution) than plasma, the required dose is administered in a smaller volume. The available evidence suggests that fibrinogen concentrates are well tolerated and can rapidly and safely restore hemostasis in patients with fibrinogen deficiencies [2–5].

Human fibrinogen concentrate (FCH), manufactured by CSL Behring (Marburg, Germany; Haemocomplettan P in Germany, and RiaSTAP in the USA), is a highly purified, lyophilized fibrinogen (coagulation factor I) powder manufactured from human plasma. All plasma used to produce FCH undergoes a viral inactivation process by means of heat treatment followed by two glycine precipitation steps [5].

Current knowledge of the pharmacokinetics of fibrinogen concentrate has been obtained from only two studies, with a sample size of five patients each [6,7]. The first, a retrospective, open-label study in patients with severe congenital hypofibrinogenemia, evaluated the pharmacokinetics of a single dose of FCH [6]. The second, a prospective, open-label study in patients with congenital afibrinogenemia evaluated fibrinogen antigen and activity as well as coagulation parameters for 14 days after a single dose of 60 mg kg−1 body weight (b.w.) of a human fibrinogen concentrate [7]. A rigorous pharmacokinetic (PK) characterization of fibrinogen concentrate in a larger sample size has not yet been reported.

The prospective, open-label study reported here provides a validated PK characterization of FCH in 14 patients with congenital afibrinogenemia, a relatively large population for this rare disease. This study also evaluated clot integrity [by measuring maximum clot firmness (MCF, mm), a functional parameter obtained using rotational thromboelastometry] and the safety of FCH.

Materials and methods

This multinational, multicenter, prospective, open-label, uncontrolled phase II study of FCH (Haemocomplettan P and RiaSTAP; CSL Behring) enrolled patients with afibrinogenemia between July 2007 and May 2008 at 10 centers in the USA and Italy. The study protocol was approved by independent ethics committees at each center, and written informed consent was obtained from each patient or a legal guardian (in the case of minors). The study was performed in accordance with the Declaration of Helsinki and the International Conference of Harmonization Good Clinical Practice guidelines.

Patients

Eligibility included age ≥ 6 years of age and undetectable plasma fibrinogen activity (i.e. < 0.2 g L−1) and antigen (i.e. < 0.004 g L−1) at screening. Exclusion criteria were as follows: bleeding disorders other than congenital fibrinogen deficiency; treatment with any fibrinogen-containing product in the preceding 2 weeks; history of deep vein thrombosis, pulmonary embolism or arterial thrombosis in the preceding year; history of esophageal variceal bleeding; acute bleeding; end-stage liver disease; an episode of multiple trauma within the preceding year; or suspicion of an anti-fibrinogen inhibitory antibody as indicated by previous in vivo recovery (IVR) < 0.5 mg dL−1 per mg kg−1 b.w., if available.

Dosing

Each patient received a single intravenous infusion of 70 mg kg−1 b.w. FCH on the morning of study. This dose was chosen to achieve a hemostatic concentration of fibrinogen on the basis of prior use of the product. FCH was supplied as 1 g of lyophilized powder to be reconstituted in 50 mL of sterile water for injection, resulting in a final concentration of 20 mg mL−1 for infusion. Concomitant treatment was recorded from enrollment until 13 days after infusion.

Assessments

Plasma samples for PK analysis were collected before infusion and 0.5, 1, 2, 4, 8, 24 and 48 h as well as 4, 6, 9 and 13 days after infusion. Plasma samples for evaluation of MCF were drawn before and 1 h after infusion. Safety was assessed on the basis of adverse events, changes in vital signs and physical examinations, and clinical laboratory assessments of hematology (platelet count, hemoglobin, and hematocrit), biochemistry (aspartate aminotransferase, alanine aminotransferase, γ-glutamyltransferase, alkaline phosphatase, total bilirubin, creatinine, urea, potassium, sodium, calcium, and chloride) and markers for coagulation activation (prothrombin fragments 1 and 2). Laboratory assessments were performed on blood samples drawn before, 1 and 24 h after and 4 and 13 days after infusion. Virus safety assessments included testing for human immunodeficiency virus (HIV-1 and HIV-2), hepatitis viruses A, B, and C and parvovirus B19 in blood samples that were collected before as well as 9 and 44 days after infusion.

Pharmacokinetics

Pharmacokinetics of FCH were analyzed for fibrinogen activity and antigen using a non-compartmental model. Fibrinogen activity was determined using a modified Clauss assay (detection limit of 0.2 g L−1), and fibrinogen antigen was determined in a fibrinogen-specific enzyme-linked immunosorbent assay (detection limit of 0.004 g L−1) using paired antibodies for fibrinogen antigen. Fibrinogen levels below the limit of detection were considered to be zero. All fibrinogen determinations were performed on frozen plasma samples at the central laboratory, CSL Behring (laboratory of U. Kalina). The following PK variables were derived individually, using standard formulae (built in to WinNonlin 5.2): initial values (within 4 h) were maximum concentration (Cmax), incremental IVR (maximum fibrinogen increase in plasma per mg kg−1 administered, after the end of infusion), and classic IVR (maximum fibrinogen increase in plasma, per mg mL−1 plasma volume dosed); terminal elimination half-life (t1/2); area under the curve (AUC) standardized to the scheduled dose of 70 mg kg−1 b.w. (AUC70 mg kg−1); clearance (CL); mean residence time (MRT); and volume of distribution at steady state (Vss). PK variables were derived using the exact dose of FCH per kg b.w. for each patient.

Clot integrity

MCF was assessed to demonstrate significant increases in clot integrity 1 h after infusion as compared with baseline. MCF was measured in the central laboratory using the rotational thromboelastometry analyzer (ROTEM; Pentapharm GmbH, Munich, Germany); this is similar to the FIBTEM test using citrated blood [8,9]. All tests were performed in duplicate. The test method was fully validated for parameters of accuracy, precision, repeatability, intermediate precision, specificity, linearity, range, and robustness. Coagulation was triggered by the tissue factor-dependent pathway. The laboratory normal range for fibrin/fibrinogen polymerization was based on plasma preparations of 50 single donations from adults, for which the mean value was 21.2 mm, with a range of 14–30 mm.

Statistical analysis

PK parameters and IVR were summarized descriptively. Descriptive statistics included median, maximum and minimum values. The change in MCF (measured in millimeters) was analyzed using a two-sided, one-sample t-test for paired observations, with a maximum permitted type I error (α) of 5%. Correlation analyses were performed for fibrinogen activity vs. antigen levels. The data were analyzed using SAS/Unix Version 9.1 (SAS Institute Inc., Cary, NC, USA) and WinNonlin Version 5.2 (Pharsight Corp., Mountain View, CA, USA).

The intention-to-treat (ITT) population included all subjects who received any portion of the FCH infusion and who fulfilled all inclusion criteria. The PK analysis per protocol (PK PP) population included all subjects in the ITT population who received ≥ 90% of the planned total dose of FCH, did not receive any fibrinogen containing blood products between infusion of FCH and the end of the 14-day PK observation period, and provided sufficient PK data to allow for a reliable PK analysis (i.e. with fibrinogen activity data up to at least 9 days postinfusion, no more than two missing consecutive fibrinogen activity values, a preinfusion fibrinogen activity value, and a maximum increase of ≥ 0.5 g L−1 in plasma fibrinogen within 4 h postinfusion).

PK analyses were performed using the PK PP population. MCF was analyzed using the ITT population.

Results

Study population

The study included 15 patients, five female (33.3%) and 10 male (66.7%). The population was 86.7% Caucasian, and the mean age was 30 years (range: 8–61 years); 11 patients (73.3%) were aged 16 to <65 years, and four patients (26.7%) were aged <16 years (range: 8–14 years). The population for which the PK parameters were derived included 14 patients because, for one patient, all plasma samples thawed during transport and were not evaluable.

Dose of FCH administered

The mean actual dose administered was 77.0 mg kg−1 b.w., although the intended dose was 70 mg kg−1, because the lot used contained 22 mg mL−1 fibrinogen rather than the nominal 20 mg mL−1 targeted.

Fibrinogen activity

The mean and median fibrinogen plasma activity levels reached a maximum within 1 h after infusion (1.33 and 1.30 g L−1, respectively), decreased continuously thereafter, and were below the limit of detection (< 0.20 g L−1) 9 days after infusion (Fig. 1A). There was no relevant difference in median (or mean) maximum concentration between patients < 16 years of age and older patients; fibrinogen plasma activity decreased to the limit of detection after 6 days for patients aged < 16 years, whereas in older patients the plasma levels returned to the limit of detection 9 days after infusion (data not shown).

Figure 1.

 Median (25% quartile to 75% quartile) fibrinogen plasma activity (A) and antigen levels (B) over time. Squares represent median values, and bars represent the 25–75% quartiles.

Median exposure for fibrinogen activity in terms of Cmax and AUC70 mg kg−1 was 1.3 g L−1 and 126.8 h mg mL−1, with a median t1/2 of 77.1 h (Table 1). PK parameters were slightly different for males and females (Table 2). Whereas Cmax, CL, and Vss were similar for both genders, t1/2 was longer for males (82.9 h) than for females (69.3 h), as were the AUC70 mg kg−1 (141.0 h mg mL−1 vs. 123.8 h mg mL−1) and MRT (104.1 h vs. 83.0 h). However, these differences were not statistically different (P > 0.1, using a two-sided, two-sample t-test). Although the numbers are small, when patients < 16 years (N = 4) were analyzed by age, they had statistically significantly (P < 0.05) higher median CL values and lower AUC70 mg kg−1, t1/2 and MRT values than patients aged 16 to <65 years (N = 10) (Table 2).

Table 1.   Pharmacokinetic (PK) parameters and in vivo recovery for fibrinogen activity and antigen (PK analysis population, N = 14)
ParameterMedian (range)
Fibrinogen activityFibrinogen antigen
  1. AUC70 mg kg−1, area under the concentration–time curve standardized to the scheduled dose of 70 mg kg−1 body weight; b.w., body weight; CL, clearance; Cmax, maximum concentration within 4 h; IVR, in vivo recovery; MRT, mean residence time; N, maximum number of subjects with available data; t1/2, terminal elimination half-life; Vss, volume of distribution at steady state.

t1/2 (h)77.1 (55.73–117.26)88.0 (54.00–116.45)
Cmax (g L−1)1.3 (1.00–2.10)1.3 (0.98–1.89)
AUC70 mg kg−1 (h mg mL−1)126.8 (81.73–156.40)122.4 (78.57–147.67)
Extrapolated part of AUC (%)7.8 (6.13–12.14)7.5 (0.82–16.13)
CL (mL h−1 kg−1)0.55 (0.45–0.86)0.57 (0.47–0.89)
MRT (h)85.9 (66.14–126.44)124.1 (87.92–170.28)
Vss (mL kg−1)52.7 (36.22–67.67)67.6 (57.74–109.16)
Incremental IVR (mg dL−1 increase per mg kg−1 b.w.)1.7 (1.30–2.73)1.7 (1.28–2.45)
Classic IVR (%)61.8 (52.45–97.43)62.3 (45.63–79.94)
Table 2.   Pharmacokinetic (PK) parameters for fibrinogen activity by gender and age
ParameterMedian (range)
Male (N = 9)Female (N = 5)< 16 years (N = 4)≥ 16 to <65 years (N = 10)
  1. *The difference between age groups was statistically significant (P < 0.05) using a two-sided, two-sample t-test. AUC70 mg kg−1, area under the concentration–time curve standardized to the scheduled dose of 70 mg kg−1 body weight; CL, clearance; Cmax, maximum concentration within 4 h; MRT, mean residence time; N, maximum number of subjects with available data; t1/2, terminal elimination half-life; Vss, volume of distribution at steady state.

t1/2 (h)82.9 (61.33–117.26)69.3 (55.73–79.45)69.1 (61.85–79.45)82.4 (55.73–117.26)
Cmax (g L−1)1.3 (1.10–1.60)1.3 (1.00–2.10)1.4 (1.00–2.10)1.3 (1.10–1.60)
AUC70 mg kg−1 (h mg mL−1)141.0 (81.73–156.40)123.8 (85.83–130.13)91.4* (81.73–130.13)135.4* (112.58–156.40)
CL (mL h−1 kg−1)0.50 (0.45–0.86)0.57 (0.54–0.82)0.77* (0.54–0.86)0.52* (0.45–0.62)
MRT (h)104.1 (66.14–126.44)83.0 (67.33–95.05)74.7* (66.14–82.98)99.6* (76.88–126.44)
Vss (mL kg−1)54.1 (48.11–59.22)47.8 (36.22–67.67)57.9 (36.22–67.67)51.4 (45.10–59.11)

Fibrinogen plasma antigen

Mean and median fibrinogen plasma antigen levels reached a maximum within 30 min after infusion (1.30 and 1.34 g L−1), and decreased continuously thereafter to 0.11 g L−1 (mean and median) 13 days after infusion (Fig. 1B). Results were similar for males and females (data not shown). When analyzed by age, the mean and median fibrinogen plasma antigen levels were similar between subgroups within the first 48 h after infusion. Starting 4 days after infusion, means and medians tended to be higher for patients aged ≥ 16 to 65 years than for patients aged <16 years (data not shown).

Median exposure for fibrinogen antigen in terms of Cmax and AUC70 mg kg−1 was 1.3 g L−1 and 122.4 h mg mL−1, with a median t1/2 of 88.0 h (Table 1). There were no relevant differences between males and females. When analyzed by age, median Cmax was similar in the two age groups (Table 2). The median t1/2, AUC70 mg kg−1 and MRT were lower for patients aged <16 years than for those aged ≥ 16 to 65 years, whereas median Vss and CL were higher for patients aged <16 years, although these results did not achieve statistical significance (data not shown).

In vivo recovery

The median incremental IVR was an 1.7 mg dL−1 increase per mg kg−1 b.w. for fibrinogen activity and antigen (Table 1). When IVR was analyzed by gender and age, no statistically relevant differences were found (data not shown).

Correlation analysis

A correlation analysis comparing fibrinogen activity with fibrinogen antigen levels showed a strong correlation between the two (correlation coefficient of 0.97 after adjustment for subject effects by univariate covariance analysis).

MCF

The difference in the mean change in MCF from before infusion to 1 h after infusion (8.9 mm) was statistically significant (P < 0.0001) (Table 3). MCF was zero at baseline for all patients, and increased to between 6.5 and 16.5 mm at 1 h after infusion of FCH.

Table 3.   Maximum clot firmness in millimeters
Time pointNMedian (range)Q25Q75P-value*
  1. *Two-sided P-value from one-sample t-test. The mean change was set to zero for two patients with missing maximum clot firmness data. Q25, 25% quartile; Q75, 75% quartile.

Before infusion130 (0–0)00
1 h after infusion1310.0 (6.5–16.5)8.512.0
Mean change159.5 (0–16.5)7.012.0< 0.0001

Safety

Two patients experienced four treatment-emergent adverse events (TEAEs), none of which was serious or led to discontinuation from the study. All TEAEs were mild in intensity, and not related to study drug. Only one TEAE (headache) occurred within 72 h after the infusion; the other three (gastroesophageal reflux disease, pain, and epistaxis) occurred at least 9 days after infusion. There were no clinically relevant changes in any of the laboratory parameters measured during the study, and nor were there any consistent or clinically relevant changes in vital signs or physical examinations during the study. There were no clinical signs of thromboembolism. All but one subject had levels of prothrombin fragments 1 and 2 that were normal or elevated preinfusion, and remained normal or normalized thereafter. There was no evidence for viral transmission.

Discussion

The available evidence suggests that fibrinogen concentrates rapidly and safely restore hemostasis in patients with congenital fibrinogen deficiency [2–5]. Their rapid onset of action, dosing flexibility, small infusion volume and ease of administration relative to other options support the use of fibrinogen concentrates as the treatment of choice in congenital fibrinogen deficiencies.

This is the first study to characterize the PK profile of fibrinogen concentrate on the basis of the use of validated assays in samples from a relatively large population (14 patients) with afibrinogenemia, the rarest and most severe form of congenital fibrinogen deficiency. The study showed that, 1 h after administration of FCH, median fibrinogen plasma activity and antigen levels rose to 1.3 g L−1, well within the recommended target of 1–1.5 g L−1 for these patients [1]. PK findings after administration of FCH were similar for fibrinogen activity and antigen. A small but insignificant effect of gender was seen on the PK parameters, with similar recovery but slightly decreased t1/2 in females. A mechanism for increased fibrinogen clearance in women is unknown. As has been seen with other plasma protein replacement products, plasma recovery in patients < 16 years old (N = 4) was similar to that in patients aged 16 to < 65 years (N = 10), but the t1/2 and AUC were shorter, with an increased Vss and Cl, which may translate into a need for more frequent dosing in children. However, the small sample size limits interpretation of this difference. The median incremental IVR was a 1.7 mg dL−1 increase per mg kg−1 b.w. for fibrinogen activity and antigen. These findings are consistent with those reported in previous PK studies in five patients [6,7], and confirm the rapid increase of fibrinogen to clinically meaningful plasma levels (i.e. median levels of 1.3 g L−1) after administration of FCH.

The clot integrity findings support the PK findings. The ability to restore clot formation was shown in this study by using the surrogate endpoint of MCF, which increased 1 h after infusion of FCH by a mean of 8.9 mm from a baseline of 0 mm (P < 0.0001). This increase was consistent with the observed increase in fibrinogen levels and the reported correlation between MCF and plasma fibrinogen levels [9], and is in the range of what is considered to be a sufficient level of fibrin/fibrinogen polymerization [8]. The improved clot formation in this study is also consistent with the efficacy seen following treatment with FCH in a previous phase IV study [4], where hemostatic efficacy was demonstrated for 26 of 26 bleeding episodes and 10 of 11 surgical interventions, and no bleedings were reported in any of 90 prophylactic infusions. Similar efficacy was reported in a clinical survey of 100 patients receiving treatment for congenital fibrinogen deficiency (target plasma fibrinogen levels: 1.0 g L−1 for minor bleeding and 1.5 g L−1 for major bleeding). [1].

This study also supports the tolerability profile of FCH. Previous studies have shown that fibrinogen concentrate has a low risk of adverse events, including allergic reactions, thrombotic events, transfusion-related lung injury, and renal toxicity [10]. Consistent with this, there were no serious TEAEs in this study, and no TEAEs that led to study discontinuation. All TEAEs were mild, and none was related to the study drug. There were no clinically relevant changes in any laboratory parameters measured during the study, including markers for coagulation activation and viral markers.

The good tolerability and safety profile was also confirmed by postmarketing surveillance data. Between 1 January 1986 and 31 August 2008, 785 366 g of FCH were distributed. A range of ‘dose per administration’ was estimated as between 1 and 8 g, depending on patient size, severity of patients’ fibrinogen deficiency, and the severity of bleeding. During this period, the manufacturer received 42 voluntary reports of suspected adverse events, representing one reported adverse event for between 2337 (for an 8-g dose) and 18 699 (for a 1-g dose) therapeutic administrations. On the basis of this postmarketing surveillance data and process validation, FCH has been shown to provide an excellent safety profile, including a high margin of virus safety.

In summary, the PK profile of FCH was robustly analyzed in a relatively large population of 14 patients with afibrinogenemia. Considering that a correlation has been shown between MCF and plasma fibrinogen levels [9], that consistent efficacy has been seen in a clinical survey [1] and in a previous efficacy study [4], and that clot integrity was increased in the present study, there is evidence that the plasma levels of fibrinogen achieved when FCH is infused at a dose of 70 mg kg−1 b.w. are sufficient to improve hemostasis with a good safety profile.

Addendum

M. J. Manco-Johnson was the principal investigator. D. DiMichele contributed to patient enrollment and creating the protocol. F. Peyvandi was coordinator for the Italian investigators and, together with P. Manucci, was also involved in enrollment and in the planning of the protocol. G. Piseddu and G. Castaman both contributed to patient enrollment. All US and Italian investigators performed the local laboratory measurements. S. Knaub was CSL Behring’s study coordinator and responsible program director.

S. Fremann was CSL Behring’s study manager, and was responsible for writing the protocol and coordinating the evaluation and clinical judgement of the results. U. Kalina was responsible for the central laboratory, ensured comparable quality of the local laboratories.

Acknowledgements

We thank the participating subjects as well as the investigators and their staff for their valuable efforts in completing this study. We also thank our friends and colleagues for various forms of assistance that led to the successful completion of this study.

Disclosure of Conflict of Interests

The study was sponsored by CSL Behring GmbH. S. Fremann and U. Kalina were employees of CSL Behring at the time when the study was performed and during manuscript preparation.

Ancillary