Factor XIII

July 19, 2002
14:00 to 18:00
Stanbro Room
Boston Park Plaza Hotel

Chairman:  A. Ichinose, Japan
Co-chairs: RAS Ariens, UK; P. Bishop, USA; C. S. Greenberg, USA; L. Muszbek, Hungary

The session was opened with apologies of absence from Drs. Ichinose, Bishop, Greenberg and Muszbek. Drs. Ariens (UK) and Seitz (Germany) resided as Acting Chairs.  There were approximately 75 people present and the presentations were followed by lively discussions.

1. Dr. Kohler (University of Bern, Switzerland) "Determination of FXIII activity using different methods. Influence of FXIIIA Val34Leu" .
    This presentation was opened with a reminder that there is no clear perspective on which type of FXIII activity assay is the best to use for any given situation. Two particular assays were discussed (i) biotin incorporation assay (PefaKit by PentaPharm) and (ii) photometric assay (Berichrom). Genetic polymorphisms of FXIII such as the Val34Leu polymorphism affect the rate of FXIII activation. The Leu allele increases the rate of FXIII activation and this may be detected when using the PefaKit assay, but will only be detected with the Berichrom assay if low levels of thrombin are used to activate FXIII. When FXIII is fully activated there is no longer any discrimination in specific activity measured with genotype variation. The question therefore is how much thrombin is generated in vivo? Previous work by Kohler et al, Thrombosis and Haemostasis, 1995, has shown that low levels of thrombin are likely to be generated (approximately 0.1u/ml) in vivo, and that there is only partial activation of FXIII. Thus, the best assay to achieve high sensitivity that can discriminate for FXIII polymorphisms is a biotin incorporation assay using low levels of thrombin to activate FXIII. To determine antigen levels of FXIII, an ELISA approach may be taken; alternatively it is possible to achieve full activation of FXIII using high levels of thrombin concentration and then assessing FXIII activity using either a photometric or biotin incorporation assay.

2. Dr. Jennings (University of Sheffield, UK) "Performance and variability of FXIII screening".
    Dr. Jennings was representing the UK NEQAS for blood coagulation screening which is currently using approximately 150 laboratories to screen for FXIII assays. Currently only approximately 20-30 laboratories use more direct FXIII assays but the majority of labs (approximately 120-130) use the clot solubility test to assess FXIII levels. In theory the clot solubility test is meant to be sensitive to approximately 1% of FXIII in plasma. The laboratories involved were given 3 plasma samples (i) normal plasma (ii) FXIII deficient plasma and (iii) FXIII deficient plasma at trough level from a patient that was about to receive prophylactic treatment. Each of the laboratories reported the level of FXIII measured and how they interpreted their result. There was a great variation of levels reported, particularly in the sample from the FXIII deficient patient at trough level (sample iii) where 70% of the laboratories classified this patient in the "normal" category. It was found that labs using a calcium-based clot solubility assay misclassified the result as normal whereas those using thrombin in their assay detected low levels. Other parameters used in the clot solubility test that affect measurement of FXIII were the lysing reagents urea or acetic acid. In May 2002 a further screening exercise was performed.  The samples were (i) FXIII deficient patient (prior to treatment at trough level) plasma taken into citrate and (ii) a pool of normal donor plasma taken into both EDTA and citrate. Again there was great variation in levels reported from the laboratories. It has been found that many variables influence the reported FXIII level; these include calcium, thrombin preparation, urea, volume of plasma/reagents, incubation time, the source of the method and normal range. Different FXIII assays also report variations for a given sample especially the Berichrom assay which reported a range of between 0-55U/dl of FXIII for a given FXIII deficient sample. There is clearly a need for a good standardization method for FXIII measurement as there is too much variation with the current methods employed. The conclusions drawn from this presentation were (i) thrombin-based clotting assays are more sensitive than calcium-based clotting assays but care must be taken to ensure that thrombin preparations do not contain calcium, (ii) there is a lack of accuracy and precision evident in FXIII assays, and (iii) the clinical relevance of ‘mild’ FXIII deficiency is unresolved.

3. Dr. Ariens (University of Leeds, UK) "Genetics of fibrin structure/function".
    Dr. Ariens opened with an introduction to FXIII and fibrinogen and showed the vast number of polymorphisms present in both FXIII and fibrinogen. Two polymorphisms of FXIII, Val34Leu in the A-subunit and His95Arg in the B-subunit, alongside two polymorphisms of fibrinogen, AaThr312Ala on the alphaC domain of the alpha-chain and BbArg448Lys on the beta-chain of fibrinogen, were expanded upon in greater detail. These polymorphisms to a certain degree affected fibrin structure (with the exception of His95Arg which was not examined). Fibrinogen level also plays a role in the outcome of fibrinogen structure, so there is an environmental effect also acting on the genotype effect. Fibrin structure was also shown to be affected (compared to controls) in plasma from first-degree relatives of patients with 2- or 3-vessel coronary artery disease. The heritability of fibrin structure was examined and showed 39% heritability of permeation(ks) in a twin study. This is lower than that observed from the heritability of fibrinogen levels (approximately 50%) and other coagulation zymogens (61-75%). This has been attributed to environmental factors which have been shown to affect fibrin structure (e.g dimethylbiguanide used for treatment of diabetes, calcium ions and ionic strength and thrombin). The conclusions of the presentation were that genetic polymorphisms affect cross-linking and fibrin structure and that environmental effects also play a role.

4. Dr. Barrowcliffe (NIBSC, UK) " Pilot study for standardization; presentation and discussion".
    Dr. Barrowcliffe introduced the process for preparing standards
(1) Preliminary investigation of materials
(2) Trial fills
(3) Stability studies and assays on trial fills
(4) Large-scale fills
(5) International collaborative study
(6) Report to WHO
Phases 1-3 have now been completed and the organization of phase 4 is beginning. The types of materials assayed for FXIII standards are plasma, concentrates, fibrin sealants (fibrinogen components) and recombinant FXIII is also being considered. In the SSC meeting last year Dr. Barrowcliffe reminded us of the stability of two FXIII concentrates which were ampouled. Preparation B was shown to have a stability of 0.05% per year at -20ºC, which was better than preparation A; therefore, further work will continue with preparation B. Proposals for the materials to be used include: plasma (NIBSC and normal pool), concentrate (one or two preparations), fibrinogen concentrate and recombinant FXIII (although this may need further consideration). Participants will be manufacturers of preparations, national control labs and clinical/academic labs. The methods to be employed will be chromogenic and antigen assessment. A preliminary collaborative study has begun with four laboratories with the aims of comparing dilution of samples in FXIII deficient plasma with buffer, comparing the Dade and Pentapharm assays, and to look at studies using fibrinogen concentrates. The samples include two FXIII concentrates, a fibrinogen concentrate and plasma. This preliminary study will be scaled up to include 10-20 labs.  Persons wishing to collaborate with this study should please contact Dr. Barrowcliffe .

5. Dr Seitz (Paul-Ehrlich-Institut, Germany) "Preliminary Results of PEI Pilot Collaborative Study"
    Dr Seitz presented preliminary results as a participating laboratory, following on from Dr. Barrowcliffe’s introduction to the preliminary collaborative study that has started with four laboratories. The aims were to compare FXIII concentrates versus plasma that are diluted in either FXIII-deficient plasma or buffer. The samples were plasma (1U/ml of FXIII content), FXIII concentrate (30U/ml FXIII content) and FXIII concentrate (50U/ml FXIII content). Preliminary results show that pre-dilution of FXIII sample in FXIII deficient plasma gives a better result than dilution in buffer containing 1% BSA. Low levels of FXIII have been found in the fibrinogen concentrates and there is a need for a comparison of methods.

6. Dr. Seitz (Paul-Ehrlich-Institut, Germany) "Preliminary Results: Novel FXIII Assay Based on Cross-linking of Peptides"
    Dr. Seitz gave an introduction to the development of a novel assay to measure FXIIIa cross-linking ability. The concept relies on the use of two peptides that are cross-linked to each other by FXIIIa. One of the peptides contains a His-tag that may be bound to Ni-NTA, and the second peptide is conjugated with FITC to allow detection. Preliminary data show that this assay is feasible but requires further development. Dr. Seitz was interested in looking for a collaborative study with a manufacturer to aid in the development of this novel assay system.