Concern over non-confirming reactive or false positive disease marker results

Several blood bankers have expressed concern over what they believe to be a significant increase in non-confirming reactive or false positive disease marker results for one or more of the following tests:

• Anti-Hepatitis B core
• Anti-HTLV I/II
• HIV-1 p24 antigen

What has been the experience in the field since January 2002? If other blood bankers have also seen a significant rise in false positive disease marker testing results, what is your hypothesis to explain it? Has anyone who observed such an increase discussed it with your kit manufacturer, and if so, what did they say?

The following replies were submitted in response to the above set of questions:

1. A blood banker in Kansas reports that his blood center's first quarter 2002 HTLV and HIV 1 p24 reactive rates have remain unchanged compared to the first quarter of 2001. He adds that their anti-HBc rates are actually lower this year than last (0.89% versus 1.16 %). He reports a problem with the HIV1 p24 antigen assay from the perspective of failed runs, usually due to the reagent blank not being in acceptable range. This has been a chronic problem.
   
   
2. A blood banker in the central valley of California reports that since December 2001, they have been experiencing a fairly drastic increase in repeat reactive, non-confirming HIV p24 antigen results. They have had the test kit manufacturer come several times to help investigate. They have looked at sample collection and all other factors, but cannot find an answer to the increase in the disease marker testing reactive rates.
   
   
3. A blood banker in Southern California, who is affiliated with a UC campus that has an ant-eating mascot, reports an inordinately high percentage of positive donors due to anti-HBc. He asks if there a publication that states what the expected deferral of first-time donors should be, when testing first time donors for anti-HBc?
   
   
4. A blood banker in Connecticut reports that they recently (February and March, 2002) had two positive HIV 1 p24 tests on autologous donors who tested negative for HIV antibody by EIA and western blot; also, HIV RNA PCR results were 'not detected.' HIV 1 p24 neutralizations have not been done (and might not be). They have had no similar scenarios in the past 3 years.
   
   
5. A Transfusion-Transmitted Viruses laboratory technologist in the Department of Transfusion Medicine at the National Institutes of Health reports that they have a number of anti-HBc positive blood donors which they think are false positives. They do NOT report seeing an increase since January 2002. They have seen in the past, usually from about October through March, a higher number of false positive anti-HTLV results and periodically anti-HCV results (also 'seasonally'). They queried their test kit manufacturer last year about the HTLV test kits and sent them their data, including the fact that their rate of false positive HTLV results for October, 2000 through March, 2001 was double what it normally ran, but still not too high. The response from the manufacturer was that while their rate had appeared to double, it was still well within the established, published 'false positive' rate for the tests.
   
   
6. A blood banker in the Northwestern U.S. reports that his blood center has seen NO dramatic increases in RR or confirmatory rates for the three assays in question. However, he reports that they have had numerous run failures, low readings, and a slight increase in RR for anti-HBc in February. The observations were reported to the test kit manufacturer and believed to be related to defective tray cover seals. The North westerner adds that they have recently begun use of a new HTLV lot that appears to have substantially higher RR rates (0.30%) but have insufficient data to determine if this rate will continue for the whole lot number. These tests have periodic fluctuations in RR rates and confirmatory rates, sometimes due to lot to lot variability. Confirmatory rates for HIV 1 p24 and HTLV have always been poor, i.e. less than 10% of RR confirms.
   
   
7. A blood banker in the City by the Bay reports that her hospital-based small donor center (about 10,000 units/year) test all units (except autologous) that are transfused at the hospital. Since January 7, they have had 7 HTLV I/II EIA repeat reactive units. Samples are sent to the Viral and Rickettsia Laboratory for HTLV confirmation. Five of these units have had negative confirmation tests. The confirmation tests are pending for the last two units. The responding blood banker believes that this is a definite increase in the reactive rate for HTLV. She is wondering if this change in test kit performance might be related to kit production because of the large number of units tested after the September 11 attack.
   
   
8. A blood banker in Boston reports that they have had four false positive HIV-1 p24 Ag tests (borderline reactive screen/indeterminate neutralization) at the end of March 2002. Prior to that, they had not seen anything unusual. The Boston blood banker has spoken with the testing lab; they said they have actually seen an overall drop in false positives since switching test kit methods!
   
   
9. An Australian blood banker wonders if the observed change in false positive disease marker rates is related to vaccination and/or an infectious etiology. He asks if it has been a bad influenza season in the United States? He recalls some scientific papers relating increased repeat reactivity rates for cell-lysate based HIV and HTLV immunoassays. He assumes (does anyone know if his assumption is correct?) that the US-licensed HTLV assays are still based mostly on cell lysates, not recombinant antigens, thus attributing the cause of false positive test results to vaccination, particularly for influenza. He imagines the mechanism to be polyclonal B cell activation, stimulating levels of junk Ig particularly against co-purified culture cell proteins. Presumably this could also effect the competitive anti-HBc assays, although it is more difficult to devise an explanation for it perturbing the HIV 1 p24 antigen assays - are they two site immunometric or some other design. The Australian reports that they don't do donor screening for HIV p24 Ag in the antipodes. Finally, he states that the list of affected assays does not included HBsAg (which is a two site immunometric assay), or anti-HIV, which presumably uses all recombinant antigen.

Submit comments to the e-Network Forum at enetworkforum@cbbsweb.org

Ira A. Shulman, MD
CBBS e-Network Forum Senior Editor & Moderator

W. Tait Stevens, MD
CBBS e-Network Forum Editor & Moderator

Elizabeth M. St. Lezin, MD
CBBS e-Network Forum Associate Editor & Moderator