Technical Library

Technical Library

BLV Update

|

Many of our customers have considered and some have asked about published sensitivity and specificity data Bovine Leukemia Virus Antibody Test Kit, ELISA (Cat. No. 284; 1 Plate and 284-5; 5 Plate).  We publish our field validation data as required and reviewed by CVB-USDA.  The idea is to make sure that companies use third party field validation data for published specificity and sensitivity information rather than in-house data which might look more enticing to the buyer.

The field validation data for our kit contained no false positives among the 280 BLV antibody negative samples tested in 3 different laboratories.  This led to jubilation, albeit temporary, on our part because we could (or had to) legitimately claim 100% specificity for the kit.  Temporary because about the time we had the sense to think it through we also had fielded a few technical service calls which indicated that some customers had encountered samples that were borderline positive with the ELISA and AGID negative. 

If there’s one thing we should have learned in 25+ years of manufacturing immunodiagnostic tests, it’s “there’s no such thing as 100% specificity, or sensitivity for that matter.”  Well, you could get something like 95% and round it up to 100%, or 51% and round up to 100%.  Or you could get lucky like we did with a specific sample set that had no false positives.  But in the real world there is no 100% sensitivity or specificity.  We have encountered false positives with this assay subsequent to the field validations and have modified our literature with the following proviso about our 100% specificity claim:

“[100% specificity is] based on a specific sample set. However, no diagnostic test kit is always 100% specific on all sample populations. Since market introduction of our BLV kit, occasional false positives have been encountered. We therefore advise all users that when BLV prevalence is low, positive samples should be confirmed by some other method, particularly where valuable animals may be involved and/or when BLV status is used as the single criterion for disposition of animals. Whenever import restrictions do not prohibit it, VMRD will provide reference assay service for positives of high-value animals or for positives in low-prevalence situations. We are not capable of testing large numbers of samples, and therefore cannot provide this reference assay service for all positives found.”

Since the realization of this problem, we have expended considerable research effort trying to improve the specificity of this assay.  In fact we have spent more time and money post introduction on this assay than on any other we have every produced. We continue to do so and have been able to make substantive progress with the current format.  We are thankful for the patience and help provided by our customers, especially, in providing problem samples for analysis.  The best way for us to make improvements is to work with the problem samples.  In doing so, we have been able to determine that anti-mouse IgG antibodies in bovine sera account for the majority of false positive reactions.  For those not familiar with the assay configuration, the gp51 antigen is captured with monoclonal mouse anti-gp51 antibody.

In a recent analysis of data from one laboratory where 2682 samples were tested by the ELISA as it currently exists, the specificity was 97.5% against AGID as the referent assay.   There were 57 samples that were determined to be “false positives.”  Of these we were able to test 43 with an improved iteration of the ELISA which registered positive on only 4.  Extrapolating and assuming all other factors would remain the same, this would have improved the specificity in that sample set to 99.9%.

We realize, as I’m sure others do, that there are no other USDA licensed BLV antibody test kits in ELISA format available for sale the United States at this time.   We are working towards licensure of the improved BLV ELISA as quickly as possible.  Our current kit is a good assay with only very rare occurrence of “false positives,” but thanks to the cooperation of our many customers and the hard work of our technicians we have an opportunity to turn a very good kit into an excellent one.  We would like to extend a “thank you” to everybody that has assisted us by providing the troublesome samples found in the field.

 

If you would like to be notified when the new format BLV kit is available for commercial release please contact VMRD, Inc. at vmrd@vmrd.com