During the last 50 years a worldwide system of standardization has been developed to ensure reproducibility of measurements in haemostasis, both in patients’ plasma and in therapeutic materials.
Summary. Accuracy and reproducibility of laboratory measurements are important in the diagnosis and treatment of bleeding disorders. This article describes the process of establishment of international standards and some of the problems that have arisen in standardization of these measurements.
During the last 50 years a worldwide system of standardization has been developed to ensure reproducibility of measurements in haemostasis, both in patients’ plasma and in therapeutic materials.
Assays of most components of the haemostatic system, and of many therapeutic materials used to treat disorders of haemostasis, are carried out on a comparative basis, relative to a standard of known potency. To relate results in one laboratory to those in other laboratories there must be some means of linking the standards used in local laboratory assays. The concept of a single biological standard that could provide such a link was first established for insulin in the early twentieth century by Sir Henry Dale . This has been developed into a well-established international system for many biological components under the auspices of the World Health Organisation (WHO).
The first International Standard (IS) in the area of haemostasis was for heparin, established in 1942 by the League of Nations, which subsequently became the WHO . In the 1960s, work commenced on establishing WHO Reference Preparations for thromboplastin reagents, because of their widespread use in control of oral anticoagulation .
This was soon followed by the establishment of the first IS for one of the clotting factors, factor VIII (FVIII) . Since then, IS have been established for most of the components of the haemostatic system. Further information about these and other standards can be found on the WHO website (www.who.int/biologicals).
For most clotting factors in plasma, the unit of activity was first defined as the amount in “average normal plasma”. When IS for clotting factors were established for the first time, they were calibrated against “average normal plasma”, to provide continuity of measurement. Once the first IS has been calibrated it is assigned a value in international units (IU), and from then on the unit of activity for that particular analyte is defined only in terms of the amount of activity in the IS. Subsequent batches of IS are calibrated in International Units against the previous standard, although there may be ongoing studies of the relationship between the IU and normal plasma, as has been the case for several of the clotting factor plasma standards (see subsequent section).
The procedure for establishment of IS has evolved over the last 50 years, and is reviewed in detail elsewhere ; a brief outline is as follows.
“Like vs. Like” A basic tenet of biological standardization is the principle of “like vs. like”, i.e. the test sample should be of similar composition to that of the standard against which it is assayed. Comparison of unlike materials, such as plasma and concentrates, tends to give high variability and differences among methods. Therefore, for most coagulation factors, IS have been established for both plasma and concentrates.
Physical attributes Certain physical requirements must be fulfilled for preparations to serve as IS. These include homogeneity (inter-ampoule variability) of the preparation and characteristics consistent with long-term stability, such as low residual moisture and oxygen content [6,7]. Homogeneity is achieved by extremely precise liquid filling. For most International Standards stability is assured by the use of sealed glass ampoules, although for some materials (e.g factor IX concentrate) stoppered vials have been found acceptable.
Collaborative study Intenational Standards undergo calibration in extensive multicentre international collaborative studies, often involving more than 20 different laboratories. Collaborative studies are planned carefully to include relevant expert laboratories (clinical, academic and commercial) and to represent the current methodologies.
Proposed assigned potencies are usually based on the consensus overall mean, and these require endorsement by study participants and by the Scientific and Standardisation Committee (SSC) of the International Society on Thrombosis and Haemostasis (ISTH) before they are submitted to the WHO Expert Committee on Biological Standardisation for formal establishment of the standard.
Stability studies The IS may be used for many years, and it is therefore essential that the preparations remain stable and the assigned values are valid for the period of use. This property is also critical for maintaining the continuity of the IU, given that replacements are calibrated relative to previous standards. Assessment of stability relies on two approaches: the accelerated degradation study and real-time stability studies, both of which are described in detail elsewhere .
Usage Because of supply limitations it is impractical to use IS as working standards in laboratories; hence the main use of International Standards is to calibrate national, regional or local standards. For plasma assays, most reagent manufacturers issue commercial plasma standards that are calibrated in International Units against the appropriate International Standard. The considerable quality control (QC) requirements for manufacturers of plasma standards could lead to excessive demand on the supply of WHO Standards, and to mitigate this, a secondary plasma standard calibrated for multiple analytes and available in large quantities, has been developed under the auspices of the ISTH; this is known as the SSC/ISTH Secondary Coagulation Standard.
Factor VIII The first IS for FVIII, established in 1971 , was a concentrate of low purity, typical of the relatively few products available at that time. It was calibrated against pools of fresh normal plasma in the 20 participating laboratories.
The variability among laboratories in this first international collaborative study was extremely high, with potencies covering a 10-fold range. Variability was somewhat lower in assays of lyophilized plasma, but this was not stable enough to qualify as an IS.
The first IS for FVIII was used successfully to calibrate manufacturers’ concentrate standards, but its use to calibrate plasma standards such as the British Plasma Standards for FVIII was less satisfactory because of high inter-laboratory variability and a 20% difference between the results of one-stage and two-stage assays . This is another example of the “like vs. like” principle: it became clear that a separate international plasma standard for FVIII would be desirable to calibrate local and commercial plasma standards. Changes in the method of collection and handling of plasma and in freeze-drying techniques led to improved stability of FVIII in lyophilized plasma, and eventually the first international plasma standard for FVIII was established in 1981 , by assay against normal plasma pools in participants’ laboratories. It was also calibrated for FVIII:Ag (previously named FVIIIC:Ag), and for von Willebrand factor antigen and activity.
Because of their high usage, both FVIII Concentrate and FVIII Plasma Standards have been replaced at fairly frequent intervals. One of the main issues has been the relationship of the IU to “average normal plasma”, which has been tested for each replacement by comparing the new standard against both the old standard and against plasma pools.
When the first IS was established, the FVIII:C content of plasma pools in participating laboratories covered a twofold range; this wide variability emphasizes both the need for an IS and also the difficulty of making this comparison.
The relationship between the IU and normal plasma has varied; in the most recent study to establish the sixth IS, the value of the sixth IS was 15% lower against the mean value of the plasma pools than against the previous IS .
Factor IX As for FVIII, a FIX concentrate standard was the first to be established by the WHO for therapeutic materials . Subsequently, an international plasma standard for FIX, together with the other vitamin K-dependent factors II, VII and X, was established by the WHO in 1987 . Most local and commercial plasma standards are now calibrated in IU.
Other coagulation factors and inhibitors The establishment of IS for the other coagulation factors and for inhibitors has followed the same pattern as for FVIII and FIX, with separate standards for plasma and concentrates, where the latter exist. Plasma standards have been established for factors II, V, VII, X, XI and XIII, VWF, fibrinogen, antithrombin, protein C and protein S. Concentrate Standards have been established for factors II, VII, VIIa and X, VWF, thrombin, fibrinogen, antithrombin and protein C.
Since the establishment of the first WHO IS for FVIII and FIX concentrates, all plasma-derived and recombinant therapeutic concentrates have been labelled in IU, where 1 IU was originally defined as the amount of analyte in 1ml of pooled, normal plasma. This approach simplifies calculations for replacement dosage and postinfusion recovery and has been remarkably successful over the last four decades.
Potency labelling for FVIII concentrates currently relies on two methods for the quantification of coagulant activity, namely, the one-stage clotting and chromogenic methods, which are preferred for product labelling in the USA and Europe respectively. The choice of FVIII potency method for labelling is irrelevant when both methods agree, but is crucial when there are significant discrepancies and the products are marketed internationally.
In the past, the labelling of such products (e.g. method-M immuno-purified and the first generation B-domain-deleted products) was managed either by maintaining formulations within the acceptable potency limits for both assay methods, or by implementing the same method for potency labelling when the product was licensed in different countries . However, when licensing authorities adopt different approaches to potency labelling, there is potential for discordance in the IU. For instance, albumin-free formulated B-domain-deleted recombinant FVIII is licensed in the USA as Xyntha (labelled by one-stage clotting assay) and in Europe as ReFacto AF (labelled by chromogenic assay), where 1 IU of the Xyntha product is equivalent to 1.38 IU of the ReFacto AF product. This example is a timely reminder of the problems we currently face with the new modified products. These products with novel properties, introduced through structural or chemical modifications (e.g. truncation, pegylation, fusion) to improve manufacturing yield or to prolong plasma half-life, will challenge the traditional approach to potency labelling relative to the WHO IS [15–17].
However, there are indications that most modified products are amenable to potency estimation using conventional methods. For instance, products of FIX fusion with albumin or the immunoglobulin Fc fraction can be measured against the WHO IS using the one-stage clotting method, and estimation of FVIII-Fc fusion molecules by both one-stage clotting and chromogenic methods has been reported, albeit with a methods discrepancy [18–20]. Potency estimation of pegylated versions of both FVIII and FIX by the one-stage clotting method appears to be associated with particular issues relating to the direct interference of the polyethylene glycol with some activated partial thromboplastin time [APTT] reagents . This is consistent with observations on pegylated FVIII, where the potency by one-stage clotting was found to be reagent-dependent (with some APTT reagents returning FVIII potency estimates as low as 10% of the expected value), whereas the chromogenic method returned expected values [22,23]. Awareness of the issue and careful choice of suitable APTT reagent has, however, allowed the one-stage clotting method to be retained for the potency estimation of pegylated FIX .
The assay behaviour of molecules, even those with similar modifications, may be difficult to predict. For instance, the B-domain-deleted FVIII molecule ReFacto AF/Xyntha has a well-characterized discrepancy between the one-stage and chromogenic methods of approximately 30%, whereas a different B-domain-deleted variant, N8, has no such difference between methods . It is possible that the length of the remaining B-domain “linker” may influence the one-stage clotting/chromogenic potency ratio . Modified therapeutics targeted towards the treatment of patients with inhibitors, such as recombinant B-domain deleted porcine factor VIII and activated FVII fused with albumin, have also been measured using conventional clotting and chromogenic methods respectively [27,28].
It therefore appears that the biological activity of most modified products can be measured in vitro using conventional methods. However, decisions on the potency labelling should be guided by a thorough characterization in vitro relative to the WHO IS, which should include the effect of different reagents (e.g. APTT reagent) and be supported by robust statistical analysis. This information should ideally be supplemented by data on activation kinetics, other techniques such as thrombin generation and elastography and, of course, in vivo studies on efficacy [19,25]. Depending on the validity of testing relative to the WHO IS, it should be possible to retain labelling in IU for some products, since the IU is defined by in vitro biological activity and does not relate to any structural or pharmacokinetic properties of the modified molecules.
Potency estimates for some modified products vary considerably when different reagents or methods are used and this will present manufacturers with various options for potency labelling [23,29,30]. However, to maintain a harmonized therapeutic approach at the global level, it is crucial that licensing authorities and manufacturers agree on the route towards the potency labelling of individual products. Once the unitage for a product has been established, this could be transferred to the manufacturer’s product reference preparation, which would restore a “like vs like” situation and resolve methods discrepancies as demonstrated previously [29,31]. Where it is not possible to obtain valid estimates in IU relative to the WHO IS, it may be necessary to label in arbitrary “product-specific units”, based on in vitro biological activity relating to product references. This strategy was previously applied to plasma-derived porcine factor VIII, which was labelled in “porcine units” .
The assay of FVIII concentrates against plasma standards has been a long-standing problem because of wide variability among laboratories and assay methods. For this reason, two separate WHO standards for plasma and concentrates were developed. However, although such comparisons are avoided in routine assays, they are relevant to manufacturers of plasma-derived concentrates, and especially to clinicians measuring in vivo recovery. In the latter situation, patients’ postinfusion samples, which essentially consist of concentrates ‘diluted’ in the patient’s haemophilic plasma, are assayed against a plasma standard.
In 1978 , it was found that when concentrates were assayed against plasma, the potencies were higher by the two-stage method than by one-stage assays – the average discrepancy from a number of collaborative studies at this time was 20%. Since then, the same trend has been found in almost every collaborative study, although the size of the discrepancy varies from study to study, and possibly with different types of concentrates.
In recent years, the chromogenic method has largely replaced the two-stage clotting method for assay of concentrates, and not surprisingly it also gives higher results than the one-stage method, being based on the same principles as the two-stage assay. A possible cause of this discrepancy may be the extensive processing applied to both plasma-derived and recombinant concentrates, which could lead to differences in their rates of activation and inactivation in the two method types from the FVIII in normal plasma; there is some evidence for this .
There is also evidence that the discrepancy is greater for recombinant concentrates than for plasma-derived products. In the collaborative study to calibrate the 5th IS FVIII concentrate, the ratio of chromogenic to one-stage potencies for a recombinant concentrate vs. the WHO plasma standard was 1.48, and in the sixth IS study  it was 1.26. These figures help to explain the large discrepancies between chromogenic and one-stage potencies found in patients’ samples after infusion of recombinant concentrates . It appears that after infusion, the recombinant products behave in an essentially similar manner in these assays to samples produced by diluting them in vitro in haemophilic plasma.
The situation with plasma-derived products is variable, depending on the nature of the product and the test systems used. For instance, in a study by Lee et al.  Hemofil M was found to give a 20% discrepancy in postinfusion plasmas between one-stage and chromogenic methods, whereas in a study of a similar product performed at CLB, there was no difference between the methods. Equivalence between the methods was also found in a UK NEQAS study on a postinfusion sample from a different type of plasma-derived concentrate.
A practical solution to this problem, which has been discussed by the FVIII/FIX Subcommittee of ISTH/SSC, is to regard the postinfusion samples as concentrates ‘diluted’ in a patient’s plasma, which is essentially what they are, and to use a concentrate standard diluted in haemophilic plasma, instead of a plasma standard, to construct the standard curve. This then provides a “like vs like” situation, and hence should provide good agreement on in vivo recoveries of recombinant concentrates when measured by chromogenic and one-stage methods. However, the nature of the concentrate standard needs to be carefully considered; it should be as similar as possible to the injected product. Thus, whereas either of the full-length recombinant concentrates could serve as a standard for the other, plasma samples following infusion of the B-domain deleted product, ReFacto, would need a Refacto concentrate standard.
This approach has been tested in in vivo recovery studies, in which patients’ samples after infusion of Recombinate, Kogenate and Alphanate were assayed against both a plasma standard and a concentrate standard. As shown in Table 1, for Recombinate and Kogenate the discrepancy between one-stage and chromogenic methods using the plasma standard was completely abolished with the appropriate concentrate standard. However, in the case of Alphanate, the use of a concentrate standard, in this case not the same as the product infused, made the situation worse. Therefore, the use of concentrate standards needs to be product specific, and should probably be restricted to recombinant and very high-purity plasma-derived products. As indicated in the previous section, this approach will probably be necessary for some of the new modified FVIII and FIX products, particularly the long-lasting pegylated molecules, because of major discrepancies between methods when compared to plasma FVIII and FIX.
|Plasma Std||Concentrate Std|
The authors stated that they had no interests which might be perceived as posing a conflict or bias.