Potency determination of factor VIII and factor IX for new product labelling and postinfusion testing: challenges for caregivers and regulators.

A workshop organized by the European Medicines Agency and the European Directorate for the Quality of Medicines and HealthCare was held in London, UK on November 28-29, 2013, to provide an overview of the current knowledge of the characterization of new factor VIII (FVIII) and factor IX (FIX) concentrates with respect to potency assays and testing of postinfusion material. The objective was to set the basis for regulatory authorities' discussion on the most appropriate potency assay for the individual products, and European Pharmacopoeia (Ph. Eur.) discussion on whether to propose revision of the Ph. Eur. monographs with respect to potency assays in the light of information on new FVIII and FIX concentrates. The workshop showed that for all products valid assays vs. the international concentrate standards were obtained and potency could be expressed in International Units. The Ph. Eur. chromogenic potency assay gave valid assay results which correlate with in vivo functionality of rFVIII products. For some modified rFVIII products and all modified rFIX products, one-stage clotting assay methods result in different potencies depending on the activated partial thromboplastin time reagent. As a consequence, monitoring of patients' postinfusion levels is challenging but it was pointed out that manufacturers are responsible for providing the users with appropriate information for use and laboratory testing of their product. Strategies to avoid misleading determination of patents' plasma levels, e.g. information on suitable assays, laboratory standards or correction factors were discussed.

Identification of kallikrein and FXIa as impurities in therapeutic immunoglobulins: implications for the safety and control of intravenous blood products.

BACKGROUND AND OBJECTIVES: The occurrence of thromboembolic events (TEEs) with intravenous immunoglobulin lots (IVIGs) raised the question of the causative agent for these adverse events. We investigated the predominant plasma proteases in 19 IVIG lots from five manufacturers including three lots associated with adverse events. MATERIAL AND METHODS: The inhibitor profile of the amidolytic activity in IVIG lots was investigated with substrates S-2302 and S-2288. In immunocapture assays, prekallikrein and FXI antigen and respective active proteases were quantified. Non-activated partial thromboplastin time (NAPTT) and a modified FXIa PTT served as global and FXIa-specific clotting assays, respectively. RESULTS: Kallikrein was identified as one major contaminant activity in IVIGs. A second activity was seen in some IVIGs with substrate S-2288, but not with S-2302. Inhibition studies excluded FXIIa, thrombin or plasmin as contaminant activity. FXI antigen was seen in all 19 IVIG lots, and FXIa activity was found as second major impurity in some IVIGs, including all lots involved in TEEs. FXIa highly correlated with a short clotting time in NAPTT. CONCLUSIONS: Kallikrein and FXIa are the major contaminants in IVIGs. FXIa was highly procoagulant, with highest level in TEE-associated IVIGs. Since the NAPTT unambiguously identified FXIa procoagulant activity in IVIGs, its implementation as a release test would improve the safety of IVIGs.

Calibration of human coagulation factor VII concentrate Ph. Eur. BRP batch 2.

The potency assay of human coagulation factor VII concentrate preparations as described in the European Pharmacopoeia (Ph. Eur.) requires a reference preparation calibrated in International Units (IU). The current Ph. Eur. Biological Reference Preparation (BRP) batch 1 was established in 2005 during an international collaborative study. It has an assigned potency of 8.2 IU/vial for the chromogenic assay method. Stocks of this BRP are dwindling and a replacement batch needs to be established. A candidate material was produced by a manufacturer from a plasma-derived concentrate preparation, with the same formulation and approximately the same potency, in the interest of continuity. The candidate material fulfilled the requirements of a BRP with regard to precision and homogeneity of fill, residual water content and stability. The potency of the candidate BRP (cBRP) was determined using chromogenic assays as required by the Ph. Eur. and in-house clotting assays in an attempt to assign a potency for both methods, as is the case for the current batch. The statistical model used for most laboratories was the maximum likelihood of the parallel line model using a logarithmic transformation of the responses. In the chromogenic assay, a potency of 9.9 IU/vial (+/- 1.8 %) was obtained for the cBRP with a very good consistency between laboratories. The results from the clotting assay, however, were less homogenous and yielded consistently higher results (13 IU/vial +/- 12 %), probably due to a higher activated factor VII (FVIIa) content than in the current BRP (3 % as compared to 0.3 %). Due to the large difference between the values obtained with the 2 different methods, it was not possible to reconcile the outcomes with each other. On the other hand, the uncertainty observed with the clotting assay method was quite large and seemed questionable for a reference preparation. Therefore the use of BRP batch 2 as a reference for the clotting assay method is not recommended. Nevertheless, the results of the study showed that the candidate BRP (cBRP) is suitable as a reference standard for the chromogenic assay according to the Ph. Eur. general chapter 2.7.10 Assay of human coagulation factor VII. It was adopted by the Ph. Eur. Commission in December 2009 as an official Ph. Eur. BRP for human coagulation factor VII concentrate with an assigned potency of 9.9 IU/vial.

Establishment of the human coagulation factor VII concentrate European Pharmacopoeia biological reference preparation batch 1.

For the potency assay of human coagulation factor VII concentrate preparations according to the European Pharmacopoeia (Ph. Eur.) a reference preparation calibrated in International Units (IU) is needed. Currently, the 1st International Standard (97/592, potency: 6.3 IU/ampoule) but no Ph. Eur. reference preparation is available. A collaborative study was run to calibrate a candidate Ph. Eur. Biological Reference Preparation (BRP) for human coagulation factor VII concentrate against the 1st International Standard; the BRP is intended to be used as working standard. A candidate BRP batch 1 was produced from a plasma-derived human factor VII concentrate preparation available on the European market. It fulfilled the requirements of a BRP with regard to precision and homogeneity of fill, residual water content and stability. In addition, the content of activated factor VII was low. Sixteen laboratories from 9 countries participated in the collaborative study. The potency of the candidate BRP was determined using the participants' chromogenic assay based on the Ph. Eur. and their in-house clotting assay, if available. The statistical model used for analysis of the results from most laboratories was the maximum likelihood of the parallel line model following a logarithmic transformation of the responses. In the chromogenic assay, a potency estimate of 8.2 IU/vial (+/-3.7%) was obtained for the candidate BRP. Results from the clotting assay were lower and less homogenous (6.7 IU/vial+/-11.6%). The results from the collaborative study showed that the candidate BRP is suitable as a reference standard for the chromogenic assay according to the Ph. Eur. It was adopted by the Ph. Eur. Commission in March 2006 as official Ph. Eur. BRP for this purpose.

The hyaluronan-binding protease upregulates ERK1/2 and PI3K/Akt signalling pathways in fibroblasts and stimulates cell proliferation and migration.

The hyaluronan-binding protease (HABP) is a serine protease in human plasma which is structurally related to plasminogen activators, coagulation factor XII and hepathocyte growth factor activator. It can in vitro activate the coagulation factor FVII, kininogen and plasminogen activators. The present study was initiated to gain a more complete picture of the cell-associated activities of this fibrinolysis-related protease. Treatment of lung fibroblasts with HABP lead to a rapid activation of signalling pathways, including the mitogen-activated protein kinase (MAPK) pathway with c-Raf, MEK and ERK1/2. Additionally the activation of the PI3K/Akt pathway and of several translation-related proteins was found. Proliferation assays confirmed the assumption of a strong growth-stimulating effect of HABP on human lung and skin fibroblasts. Intracellular signalling and growth stimulation were strongly dependent on the proteolytic activity of HABP. Stimulation of signalling and proliferation by HABP involved the fibroblast growth factor receptor 1 (FGFR-1). HABP-stimulated proliferation of lung fibroblasts MRC-5 was accompanied by a significant intracellular increase in basic fibroblast growth factor (bFGF), the major ligand of FGFR-1; bFGF could however not be identified in the supernatant of HABP-treated cells. Though, the conditioned medium from HABP-treated cells showed a strong growth-promoting activity on quiescent fibroblasts, indicating the release of a yet unknown growth factor amplifying the initial growth stimulus. In a two-dimensional wound model HABP stimulated the invasion of fibroblasts into a scratch area, adding a strong pro-migratory activity to this plasma protease. In summary, HABP exhibits a significant growth factor-like activity on quiescent human lung and dermal fibroblasts. Our findings suggest that this fibrinolysis-related plasma protease may participate in physiologic or pathologic processes where cell proliferation and migration are pivotal, like tissue repair, vascular remodelling, wound healing or tumor development.

Tissue factor is the only activator of coagulation in cultured human lung cancer cells.

It is a long-known principle that tumour cells tend to exploit the host's physiologic systems in order to get support in terms of, for example, nutrition, growth or metastasis. One of these physiologic systems is the blood coagulation cascade, which has been found activated in many tumour patients. The mechanisms of the activation of coagulation have been assessed in numerous animal and in vitro experiments, and the results appeared to point to several distinct activators. The present study used a large panel of different cultivated human lung cancer cell lines and experimental systems involving normal plasma, plasmas deficient of factors V, VII or X, purified coagulation factors II and X, recombinant tissue factor (TF), and specific inhibitory antibodies against factor VII and TF. The results provide strong evidence that there is no activator of coagulation besides TF in the wide array of lung cancer cells examined. However, this work reveals a striking variability of TF content among the cell lines. This might explain ambiguous results of clinical trials of anticoagulation as an adjunct to antineoplastic therapy in lung cancer. By sensitive diagnostic tools like the plasma thrombin-antithrombin complex levels it might be possible to select patients with activated coagulation, who might benefit from anticoagulation.

Activation of proPHBSP, the zymogen of a plasma hyaluronan binding serine protease, by an intermolecular autocatalytic mechanism.

The hyaluronic acid binding serine protease (PHBSP), an enzyme with the ability to activate the coagulation factor FVII and the plasminogen activator precursors and to inactivate factor VIII and factor V, could be isolated from human plasma in the presence of 6M urea as a single-chain zymogen, whereas under native conditions only its activated two-chain form was obtained. The total yield of proenzyme (proPHBSP) was 5-6 mg/l, corresponding to a concentration of at least 80-100nM in plasma. Upon removal of urea, even in the absence of charged surfaces a rapid development of amidolytic activity was observed that correlated with the appearance of the two-chain enzyme. The highest activation rate was observed at pH 6. ProPHBSP processing was concentration-dependent following a second order kinetic and was accelerated by catalytic amounts of active PHBSP, indicating an intermolecular autocatalytic activation. Charged macromolecules like poly-L-lysine, heparin, and dextran sulfate strongly accelerated the autoactivation, suggesting that in vivo proPHBSP activation might be a surface-bound process. The intrinsic activity of the proenzyme was determined to be 0.25-0.3%, most likely due to traces of PHBSP. The presence of physiological concentrations of known plasma inhibitors of PHBSP, like alpha2 antiplasmin and C1 esterase inhibitor, but not antithrombin III/heparin, slowed down zymogen processing. Our in vitro data suggest that the autoactivation of proPHBSP during plasma fractionation is induced by the removal of inhibitors of PHBSP and is accelerated by charged surfaces of the chromatographic resins.

Virus safety of prothrombin complex concentrates and factor IX concentrates.

Substantial progress in virus safety has been achieved during the past 15 years. Therefore only a few virus transmissions with plasma-derived products have been observed since 1985. Specific steps to eliminate, remove, or inactivate viruses were developed. Although virus safety is of decisive importance, chemical or physical treatment during the manufacturing process need not activate labile coagulation factors that cause the risk of thrombogenicity nor need not create neoantigens that mediate the risk of inhibitor formation. Thus, any new virus elimination procedure has to be evaluated and validated for all safety aspects. A comprehensive framework of regulations and efforts has been set up involving plasma donors, donation centres, manufacturers, regulatory authorities, politicians responsible for legislation, physicians, and patients. Blood and plasma donation centres and pharmaceutic industry follow "Good Manufacturing Practice" and are subject to regular audits and official inspections. Every single donation as well as plasma pools are tested for virus markers. The final products need both a marketing authorization and official batch release; in Germany, supervised by the Paul-Ehrlich-Institute. European integration is the purpose of the European Medicines Evaluation Agency. An alert pharmacovigilance system enables scientifically adequate reactions in any case of a safety problem. The ultimate evidence of product safety is provided by clinical surveillance. By participating in clinical studies, patients themselves are able to contribute significantly to the safety of plasma-derived products. The currently achieved high level of safety should encourage us to take further steps to stabilize this success and to look for further progress, wherever possible.

Detection of a novel plasma serine protease during purification of vitamin K-dependent coagulation factors.

A novel serine protease (PHBSP) was purified from human plasma by two chromatographic steps with a final yield of 1.6 mg/l plasma. The protease consists of two disulfide-bridged chains of about 50 and 30 kDa with the light chain containing the active site of the enzyme. NH2-terminal sequence analysis revealed identity to the deduced amino acid sequence of HGFA-like mRNA. The activity of PHBSP is strongly dependent on Ca2+ ions and is efficiently inhibited by alpha2-antiplasmin and aprotinin. Possible functions of PHBSP in the hemostatic system are discussed.

Identification of the 170-kDa melanoma membrane-bound gelatinase (seprase) as a serine integral membrane protease.

The 170-kDa membrane-bound gelatinase, seprase, is a cell surface protease, the expression of which correlates with the invasive phenotype of human melanoma and carcinoma cells. We have isolated seprase from cell membranes and shed vesicles of LOX human melanoma cells. The active enzyme is a dimer of N-glycosylated 97-kDa subunits. Sequence analysis of three internal proteolytic fragments of the 97-kDa polypeptide revealed up to 87.5% identity to the 95-kDa fibroblast activation protein alpha (FAPalpha), the function of which is unknown. Thus, we used reverse transcription-polymerase chain reaction to generate a 2.4-kilobase cDNA from LOX mRNA with FAPalpha primers. COS-7 cells transfected with this cDNA expressed a 170-kDa gelatinase that is recognized by monoclonal antibodies directed against seprase. Sequence analysis also showed similarities to the 110-kDa subunit of dipeptidyl peptidase IV (DPPIV). Like DPPIV, the gelatinase activity of seprase was completely blocked by serine-protease inhibitors, including diisopropyl fluorophosphate. Seprase could be affinity-labeled by [3H]diisopropyl fluorophosphate, but the proteolytically inactive 97-kDa subunit could not, confirming the existence of a serine protease active site on the dimeric form. Proteolytic activity is lost upon dissociation into its 97-kDa subunit following treatment with acid, heat, or cysteine and histidine-modifying agents. We conclude that seprase, FAPalpha, and DPPIV are related serine integral membrane proteases and that seprase is similar to DPPIV, the proteolytic activities of which are dependent upon subunit association.

Thrombin inhibitors of bloodsucking animals.

Hematophagous animals have the unique ability to inhibit blood coagulation when sucking blood from a wound. This article concerns thrombin inhibitors from these animals, specifically the one from the insect Rhodnius prolixus. To date, the most-studied inhibitor from these animals is hirudin, which specifically neutralizes thrombin but no other clotting serine proteases. The biochemical properties of hirudin are described. Hirudinlike thrombin inhibitors from the Asian leech Hirudo manillensis, Haemadipsa sylvestris, and Haemadipsidae are also discussed. In addition, a thrombin inhibitor from the insect Rhodnius prolixus, Rhodniin, is extensively reviewed. There is considerable interest in these inhibitors because they may be found useful as treatment modalities for thromboembolic disorders. Hirudin is already extensively investigated, and some of the others may follow. Potentially these new inhibitors could be of greater clinical benefit than the presently used heparins.

Structure/function implications for the aminopeptidase specificity of aleurain.

The cysteine protease aleurain, a member of the papain superfamily, was characterized by its specificity constants, kcat/Km, for the hydrolysis of different substrates of the type H-P1-NH-Mec (NH-Mec, 4-methylcoumaryl-7-amide). The determined constants for the different substrates decrease in the order citrulline > Arg = Phe >> Ala. A 75-fold loss of specificity was observed when the substrate Bz-Arg-NH-Mec (Bz, benzoyl), with a blocked N-terminus, was used instead of H-Arg-NH-Mec. The pH dependence of kcat/Km for H-Arg-NH-Mec was bell-shaped with pKa1 and pKa2 values of 5.81 and 7.27, respectively, at 25 degrees C. The residue corresponding to a pKa1 value of 5.81 could be identified by its ionisation enthalpy, delta Hion, of 15 kJ/mol as a carboxylate group of the enzyme interacting electrostatically with the residue with pKa2 7.27, attributed to the alpha-amino group of the substrate by its delta Hion value of 48 kJ/mol. Aleurain can be titrated at the active site with L-trans-epoxy-succinylleucylamido(4-guanidino)butane, and the reaction was characterized by its association rate constant of 19,000 M-1.s-1. Native chicken cystatin inhibited aleurain competitively with Ki 133 nM. Recombinant chicken cystatin variants Ala-Glu-Phe-[Met1, Ile29, Leu89] chicken egg-white cystatin, (variant 1) and the N-terminally truncated form des-(S1-P11)-[Ala12, Glu12, Phe14, Met15, Leu89]-chicken egg-white cystatin, (variant 2), inhibited aleurain competitively with Ki values of 125 nM and 5 microM, respectively. Implications for the aminopeptidase activity of aleurain are discussed using cathepsin H for comparison.

A Kazal-type inhibitor with thrombin specificity from Rhodnius prolixus.

A thrombin-specific inhibitor with an apparent molecular mass of 11 kDa has been purified from the insect Rhodnius prolixus. Amino-terminal protein sequence analysis allowed the molecular cloning of the corresponding cDNA. The open reading frame codes for a protein of about 103 amino acid residues and displays an internal sequence homology of residues 6-48 with residues 57-101 indicating a two-domain structure. Based on the amino acid sequence the two domains exhibit high homology to protease inhibitors belonging to the Kazal-type family. Model building suggests that the first domain binds to the active site with residue His10 pointing into the specificity pocket. From gel filtration and tight-binding inhibition experiments the inhibitor appears to form 1:1 complexes with thrombin. Periplasma-directed heterologous expression of the rhodniin cDNA in Escherichia coli yields the intact thrombin inhibitor. Natural and recombinant rhodniin both display inhibition constants of about 2 x 10(-13) M.

Isolation, sequence analysis, and cloning of haemadin. An anticoagulant peptide from the Indian leech.

A slow, tight-binding inhibitor of thrombin with an apparent molecular mass of about 5 kDa has been isolated from Haemadipsa sylvestris, an Indian leech of the family of Haemadipsidae. The inhibitory activity, called haemadin, is thrombin specific since it does not inhibit other proteases like trypsin, chymotrypsin, factor Xa, or plasmin. NH2-terminal amino acid sequence analysis (residues 1-45) does not reveal any homology to known serine protease inhibitors, including the thrombin-specific inhibitor hirudin. The haemadin cDNA cloned by polymerase chain reaction techniques codes for a polypeptide of 57 amino acid residues preceded by 20 residues of a signal peptide sequence. A synthetic gene coding for the mature haemadin was expressed in Escherichia coli. Recombinant haemadin displays a similar inhibition constant and specific activity as its natural counterpart. Although there is no obvious sequence identity between haemadin and hirudin, both proteins seem to share common mechanisms for thrombin inhibition.

Hirudisins. Hirudin-derived thrombin inhibitors with disintegrin activity.

Recombinant hirudin variants have been designed which inhibit alpha-thrombin by the hirudin mechanism and which in addition exhibit disintegrin activity. These proteins, called "hirudisins," have been engineered by replacing the Ser-Asp-Gly-Glu sequence at the tip of hirudin's finger-like structure (residues 32-35) by Arg-Gly-Asp-Ser (RGDS) to yield hirudisin and Lys-Gly-Asp-Ser (KGDS) to obtain hirudisin-1. Comparison of thrombin inhibition activities showed that hirudisin is 2-fold more potent (K(i) = 160 +/- 70 fM) than hirudisin-1 (K(i) = 370 +/- 44 fM) and recombinant (r)-hirudin (K(i) = 270 +/- 50 fM). alpha-Thrombin-stimulated platelet aggregation was effectively inhibited by r-hirudin, hirudisin, and hirudisin-1 with IC50 of 5.7 to 6.8 nM. Unlike r-hirudin, hirudisin inhibits ADP-induced platelet aggregation (IC50 = 65 microM) 3- to 5-fold stronger than the linear GRGDS- and RGDS-peptide. Direct interaction of hirudisin with purified glycoprotein IIb-IIIa demonstrated that antiplatelet aggregation activity is due to the integrin-directed RGD motif. Disintegrin activity of hirudisin relative to that of reduced and carboxymethylated hirudisin suggests that the conformational strain favors binding to integrins. On the basis of these results, hirudisins appear to be interesting molecules for the design of potential antithrombotic agents with antithrombin as well as antiplatelet aggregation activities.

Studies on the aminopeptidase activity of rat cathepsin H.

Three synthetic substrates H-Arg-NH-Mec, Bz-Arg-NH-Mec and H-Cit-NH-Mec (Bz, Benzoyl; NH-Mec, 4-methylcoumaryl-7-amide; Cit, citrulline) were used to characterize specificity requirements for the P1-S1 interaction of cathepsin H from rat liver. From rapid equilibrium kinetic studies it was shown that Km, kcat and the specificity constants kcat/Km are quite similar for substrates with a free alpha-amino group. In contrast, a 25-fold decrease of kcat/Km was observed for the N-terminal-blocked substrate Bz-Arg-NH-Mec. The activation energies for H-Arg-NH-Mec and Bz-Arg-NH-Mec were determined to be 37 kJ/mol and 55 kJ/mol, respectively, and the incremental binding energy delta delta Gb of the charged alpha-amino group was estimated to -8.1 kJ/mol at pH 6.8. The shown preference of cathepsin H for the unblocked substrates H-Arg-NH-Mec and H-Cit-NH-Mec was further investigated by inspection of the pH dependence of kcat/Km. The curves of the two substrates with a charged alpha-amino group showed identical bell-shaped profiles which both exhibit pKa1 and pKa2 values of 5.5 and 7.4, respectively, at 30 degrees C. The residue with a pKa1 of 5.5 in the acid limb of the activity profile of H-Arg-NH-Mec was identified by its ionization enthalpy delta Hion = 21 kJ/mol as a beta-carboxylate or gamma-carboxylate of the enzyme, whereas the residue with a pKa2 of 7.4 was assigned to the free alpha-amino group of the substrate with a delta Hion of 59 kJ/mol. Bz-Arg-NH-Mec showed a different pH-activity profile with a pKa1 of 5.4 and a pKa2 of 6.6 at 30 degrees C. Cathepsin H exhibits no preference for a basic P1 side chain as has been shown by the similar kinetics of H-Arg-NH-Mec and the uncharged, isosteric substrate H-Cit-NH-Mec. In summary, specific interactions of an anionic cathepsin H active site residue with the charged alpha-amino group of substrates caused transition state stabilization which proves the enzyme to act preferentially as an aminopeptidase.

Interaction of hirudin with the dysthrombins Quick I and II.

The interaction of hirudin with the dysfunctional enzymes thrombin Quick I and II has been investigated. Natural and recombinant hirudin caused nonlinear competitive inhibition of thrombin Quick I. The results were consistent with thrombin Quick I existing in two forms that have different affinities for hirudin. The affinities of these forms for natural hirudin were respectively 10(4)- and 10(6)-fold lower than that of alpha-thrombin. In contrast, truncated hirudin molecules lacking the C-terminal tail of the molecule caused linear inhibition of thrombin Quick I. These results indicate that different modes of interaction of the two forms of thrombin Quick I with the C-terminal tail of hirudin were the cause of the nonlinear inhibition. Comparison of the dissociation constants of thrombin Quick I with the truncated and full-length forms of hirudin suggested that the interactions that normally occur between the C-terminal tail of hirudin and thrombin were completely disrupted with the low-affinity form of thrombin Quick I. Thrombin Quick II displayed an affinity for natural hirudin that was 10(3)-fold lower than that observed with alpha-thrombin. In contrast, it bound a mutant hirudin with altered N-terminal amino acids only 16-fold less tightly. These results are discussed in terms of structural alterations in the active-site cleft in thrombin Quick II.