Fibrinogen and Factor XIII 2007 SSC Subcommittee Minutes

This was the first meeting of the combined Fibrinogen & Factor XIII subcommittee in a single session of 4 hours and some experiences were shared: A common concern is that the time for the meeting is short, and that several aspects could not be discussed with sufficient detail (especially the nomenclature and the laboratory aspects).

PART 1: Fibrinogen

In the first presentation, dr. W. Koenig showed us that the biological variation of fibrinogen levels in plasma affects the relationship between plasma fibrinogen levels and risk of cardiovascular disease. The reliability index of fibrinogen measurements is around 0.5-0.7, which means an 30-50% underestimation of the risk estimate in epidemiological studies. Multiple measurements (2-3 samples collected at least 2 weeks apart) and exclusion of samples collected during inflammatory conditions improves the risk estimation.

In the next presentation, dr. L. Medved and dr. J. Weisel gave an update on the nomenclature of the fibrinogen molecule and on the nomenclature of the fibrin formation. A major point to consider is the numbering of the amino acids, since it is common use for fibrinogen to use a numbering system, based on the mature protein, while the recommendations of the Human Genome Variation Society use the transcription inititiation site as +1. This approach is now also being discussed for other (hemostasis) proteins. Since everybody is very much used to the old numbering system, it has been suggested to use double numbering to avoid confusion. A report is being prepared which is expected to be finalized in 2008.

The next presentation was also on nomenclature, now of the fibrinogen variants, and dr. M. de Maat and dr. J. Koopman prepared and presented a suggestion for a clearer nomenclature. A report is now being prepared that will be circulated among the fibrinogen investigators and will also be finalized in 2008.

Since it was noticed last time that it is difficult to compare different publications on the characteristics of fibrinogen mutations, Dr. M. Neerman-Arbez presented a minimum set of assays that should always be performed. Of course, different types of mutations (resulting in afibrinogenemia or dysfibrinogemia) require different approaches. This list will be included in a SSC report and several investigators that were present agreed to participate in the discussions.

The next two presentations focused on the effects of measuring fibrinogen with different assays in the association with cardiovascular risk. First dr. A. Silveira showed that in the recent meta-analysis of the Fibrinogen Research Collaboration, there is no indication that the association with risk is different for the different types of assays. Also, in a recent QTL analysis, they showed that the association between genetic markers and fibrinogen levels is similar for the Clauss assay and a nephelometric assay. In the next presentation dr. D. Peetz presented first results of the Gutenberg Heart Study, in which 5 different fibrinogen assays were used, and here the correlation between the assays varied between 0.5 and 0.9. A common concern is that fibrinogen assays need to be further standardised.

PART 2: FXIII

Session I. International Standardization and Registry Issues.

A. Ichinose gave a report about the FXIII SWP business meeting in Dresden in February 2007: A) Political/Social issues: The current chair of the FXIII subcommittee explained that he initially could not accept the chair of the merged subcommittee, but was able to accept a joint chairmanship. The FXIII Standardization Working Party (SWP) decided to continue to work under the FXIII subcommittee of SSC/ISTH, and to centralize to the FXIIISWP all activities regarding the standardization of Factor XIII to ensure formal and consistent practices. B) Credit issues: On petition of the chair of the FXIIISWP, ISTH/SSC headquarters opened a homepage in their website in order to publicize/announce the major contributions of the FXIIISWP for the establishment of the 1st International standard for plasma (ISP) of FXIII. In addition, one FXIIISWP member was asked to complete a paper on the 1st ISP of FXIII. In order to avoid the recurrence of complications, the FXIIISWP decided to make written agreements for confirmation in advance and when needed. C) Financial issues: Financial difficulties were reported individually, and possibilities to raise funds for the standardization activities were discussed. The FXIIISWP decided to estimate the amounts of money required for direct research costs and travel expenses to attend the FXIIISWP meetings, and agreed to seek possible financial supporters. D) Scientific & Technical Issues: Difficulties in reproducible measurements of XIIIA were considered carefully. The FXIIISWP decided to perform a pilot study to explore possible solutions to this problem. The requirements for good FXIII-deficient plasma and antibodies against various forms of FXIII were also confirmed. E) Clinical issues: the FXIIISWP addressed the necessity of standards for plasma FXIII concentrates and rec. FXIII preparations, for clinical management of FXIII deficiency.

A Report on standardisation of FXIII Concentrate on behalf of the FXIII SWP was presented by Sanj Raut. Following ISTH/SSC FXIII & Fibrinogen Subcommittee & SWP meetings where the need for a FXIII concentrate reference standard was established, it was proposed to carry out a pilot study (activity & antigen) on all available FXIII concentrate materials. Aims of the study was to evaluate candidate materials for the establishment of the 1st IS for FXIII concentrate, to investigate the relationship between measurement of FXIII in the concentrate vs plasma and to investigate the relationship between measurement of FXIII activity and FXIII antigen levels. Trial fills & accelerated degradation stability studies of candidate materials have been completed and samples & assay kits have been shipped out to the SWP participating laboratories together with protocol and assay design. Study is currently on hold whilst a number of issues are being resolved but it is envisaged that the study will resume later in 2007. 5 materials were provided: (X) FXIII Concentrate (02/170), activity potency ~ 40 IU/amp; (Y) WHO 1st IS FXIII Plasma (02/206): activity potency 0.91 IU/amp; antigen potency 0.93 IU/amp; (Q) rFXIII Concentrate (06/021PM), activity potency ~ 40 IU/amp; (R) rFXIII Concentrate (06/022PM), activity potency ~ 40 IU/amp; (J) FXIII Concentrate J, potency ~ 40 IU/vial. Laboratories were asked to use routine and/or provided FXIII activity and antigen (A2B2-pdFXIII; A2 subunit-rFXIII) assays. They were requested to carry out 4 independent assays on each sample (on 2 separate days) using preparation Y as standard and following assay instructions/design as described in the protocol. They were to pre-dilute in FXIII deficient plasma and submit all raw data for analysis (by Nov 2007). It is envisaged that results will be analysed and presented at the SSC in June 2008. Based on this, selection of materials for definitive fills will be carried out and a full international collaborative study would be initiated. The objective is to submit study report to WHO/ECBS for establishment in Oct 2009.

V. Ivaskevicius focused on the progress of the international FXIII registry. Recently the data summarizing the former FXIII Registry of ETRO Working Party was published in Throm Haemost (2007, 97;6:914-21). A new on-line Questionnaire for patients affected by FXIII deficiency was presented. This questionnaire is available on the www.f13-database.de website. Further, Standardization of Genetic Terms were discussed in relation to FXIII genes and of clinical issues. Advantages and disadvantages were shown of new nomenclature supported by Human Genome Variation Society and old traditional nomenclature. Proposals were provided regarding classification of degree of severity depending from FXIII activity and bleeding symptoms.

Session II: Scientific and Clinical Issues:

W. Korte showed data confirming previous publications that a seemingly moderate intraoperative decrease of FXIII to levels below ca. 60% is associated with subsequent bleeding. A prospective study on intraoparative FXIII substitution was terminated early after 22 patients already, since a significant reduction of bleeding was observed. Further studies on the postoperative setting will follow.

Concerning diagnosis of FXIII deficiency, five cases were presented by H.P. Kohler, where FXIII levels had been overestimated by the most widely used Berichrom assay finding up to 15% FXIII activity despite non-detectable FXIII antigen; a problem already described in recent literature. Notably, it has been shown that overestimation can be amended by the subtraction of the blank; however manufacturer and user do not appear to be aware. L. Muszbek presented an algorithm for the laboratory diagnosis and classification of FXIII deficiencies: 1/ Screening test: A quick functional assay for the determination of plasma FXIII activity; 2/ Mixing study for the detection of neutralizing antibody; 3/ If FXIII activity is <5%, further functional test for the precise assessment of FXIII activity in the low activity range (amine incorporation assay, evaluation of fibrin cross-linking by SDS PAGE); 4/ Determination of FXIII A2B2 complex (R-ELISA); 5/ If the concentration of the complex decreased determination of individual FXIII subunits in the plasma; 6/ Determination of platelet FXIII activity and FXIII-A concentration; 7/ Detection of non-neutralizing antibodies against FXIII subunits by binding assays; 8/ Molecular genetic investigations. Guidelines on diagnosis and monitoring of FXIII therapy are missing, and it is proposed to publish an SSC position paper. This proposal was endorsed by the attendees.

Dr. Kitano discussed genetic and molecular bases of phenotypes of the B subunit of Factor XIII. Three major protein phenotypes of FXIII-B (FXIIIB*1, FXIIIB*2, and FXIIIB*3) are determined by isoelectric focusing and immunoblotting. FXIIIB*1 is the most common phenotype among Europeans, while FXIIIB*2 and FXIIIB*3 are common in Africans and Asians, respectively. FXIIIB*4 is a rare phenotype. Recently, we determined amino acid residues responsible for each phenotype by nucleotide sequencing analysis using genomic DNAs. Assuming that FXIIIB*1 would be a basic phenotype, FXIIIB*2 had an amino acid substitution of codon 95 in exon III from His to Arg, and FXIIIB*4 had an exchange of codon 368 in exon VII from Glu to Val. For FXIIIB*3, we discovered a C-to-G change in intron K. This nucleotide substitution would create a better splicing acceptor AG dinucleotide, result in differential splicing of intron K, and produce a totally new exon. The presence of this message was confirmed by RT-RCR using hepatic mRNA. As a result, FXIIIB*3 has a 15 residues-longer carboxy-terminal than other phenotypes as well as two additional basic and one extra acidic amino acid residues. Accordingly, all four phenotypes contain variable numbers of charged residues, which ultimately contribute to their differential isoelectric points.

N.T.P. Bakker spoke about the hypothesis that chronic changes in blood flow and blood pressure induce an adaptation of vascular calibre, and that this remodelling depends on the cross-linking enzyme tissue-type transglutaminase (tTG). Blood pressure-dependent and flow-dependent remodelling was studied in wild-type (WT) and tTG-null mice using a surgically imposed change in blood flow in small mesenteric arteries. WT mice showed inward remodelling after 2 days of low blood flow, which was absent in arteries from tTG-null mice. Yet, after continued low blood flow for 7 days, inward remodelling was similar in arteries from WT and tTG-null mice. Studying the alternative pathways of remodelling, we identified monocytes/macrophages as a source of factor XIII and backup mechanism in tTG null mice.

R. Ariens addressed in his presentation the role of gamma dimer formation in determining clot elasticity and lysis rates is discussed. Cross-linking of fibrin by FXIIIa occurs between gamma-gamma and alpha-alpha chains. The relative contribution of gamma-gamma chain cross-linking is poorly understood. We made mutations in the gamma chain cross-linking sites and investigated fibrin structure and function. We found that gamma-dimer formation contributes significantly to clot rigidity. Gamma-alpha hybrid cross-links did not effectively increase clot rigidity. Differences in fibrin degradation products were observed in fibrinogen cross-linking mutants. However, no differences in fibrinolysis rates were observed when gamma-gamma cross-linking was eliminated. We conclude that gamma chain cross-linking plays a major role in determining clot rigidity but not lysis.