Here is a June 6, 2008 message from Kim Kinney at Clarian in Indianapolis:
Hi George,
We are going to evaluate a kit for collagen binding. I remember at the Midwest Symposium this year Dr. Adcock commented on the type of collagen used in the kit and that one was better than another. From what we can find, there is either human or equine collagen. Is there one better than another? From our reading it seems as if the equine will bind more HMW multimers while the human binds more low and moderate weight multimers…is there a preference when it comes to this assay?
Thanks, Kim
Hi, Kim. Thank you for your question. For answers, with assistance from Steve Duff of Precision, I’ve gone to Favaloro EJ. An update on the von Willebrand factor collagen binding assay: 21 years of age and beyond adolescence but not yet a mature adult. Semin Thromb Hemost 2007;33:727-44. I’ll also direct everyone to our latest modules, von Willebrand disease parts 1 and 2. I discuss the collagen binding assay (VWF:CB) briefly in the module and will expand on it here.
Most labs currently profile von Willebrand disease (VWD) using the von Willebrand factor (VWF) antigen immunoassay (VWF:Ag), ristocetin cofactor assay (VWF:RCo), and the coagulation factor VIII (FVIII:C) activity assay. VWF:RCo measures VWF activity because the assay responds to high molecular weight VWF (HMW VWF) but not small or intermediate VWF multimers. Results are interpreted to classify VWD type 1 from type 2A and 2B. A VWF:Ag/VWF:RCo ratio >0.7 is interpreted as type 1; when ≤ 0.7, we conclude it is either type 2A or 2B.
Favaloro warns that VWF:RCo, which is a platelet agglutination test performed on an aggregometer, has no international standard, is complex and time-consuming, requires special, time-dependent specimen management; generates CV% levels of 20–40% under the best of analytical circumstances, and is insensitive to VWF activity levels below 20%. He quotes three studies that conclude our current profiles consisting of VWF:RCo, VWF:Ag, and FVIII:C misclassify 25% of types 2A or 2B as VWD type 1.
The VWF:CB immunoassay employs collagen as its solid-phase target on the principle that collagen differentially binds HMW VWF. The VWF:CB is an effort to provide a reproducible, sensitive, quantitative substitute that skirts the limitations of the VWF:RCo. However, VWF:CB has never successfully substituted for VWF:RCo, and the labs that have adopted the former have added it to the profile without dropping the latter. Currently available VWF:CB kits generate CV% values of 15–25%.
The key to VWF:CB success lies in collagen formulation. Commercial distributors currently use purified type III collagen, the flexible reticular form that supports vertebrate connective tissue. Type III collagen provides favorable target binding and substrate color development characteristics. Most type III collagen is extracted and purified from human connective tissue though some comes from bovine or equine sources.
Favaloro contends formulation using 95% type I and 5% type III collagen, though possessing less desirable assay characteristics, more specifically responds to HMW VWF. Type I collagen is extracted from bone and tendons, and the source is most often equine or bovine, though it may also be human. He demonstrates that pure type III collagen binds small and intermediate VWF multimers, though with less avidity than HMW VWF, and misclassifies 10% of VWD types 2A and 2B as VWD type 1. He concedes that pure type I collagen binds VWF poorly, but the addition of 5% type III optimizes the assay.
The original hope was the VWF:CB, simpler to manage, with superior CV% and HMW VWF discrimination would replace VWF:RCo, however a profile that exclusively uses one or the other still tends to be variable and to miss VWD type 2. Favaloro recommends screening with VWF:Ag, VWF:CB, and FVIII:C, then performing VWF:RCo when one or more of the three falls below the reference range. With this approach, he contends it is unnecessary to follow up with VWF multimeric analysis. Finally, the combination of a reduced VWF:Ag and VWF:RCo with a normal VWF:CB indicate that the VWD is subtype 2M, reduced VWF platelet binding.
I hope this helps. If the problems of VWF:CB reproducibility and specificity are resolved, it may still be the best assay for the initial VWD profile. Geo
No comments here.