werner@aecom.UUCP (Craig Werner) (03/17/85)
Dateline: last week (but I was busy): AP) The Government has approved a sec- ond screening test for blood contam- inated by acquired immune deficiency syndrome, or AIDS, after laboratory results indicating it was the most sens- itive indicator of possible exposure to the disorder. Electro-Nucleonics Inc. of Fairfield, N.J., was awarded the second license by the FDA .... About 2 months ago, I ran a series of four articles on HTLV-3, the virus that most likely causes AIDS. I spoke of the screening test as highly accurate, and then received mail and flames otherwise. Well, here is the source of the misunderstanding. There are several tests - the most sensitive being the Western blot, which I was referring to. (This was reasonable, since most Test-Positive results were to be redone with Western blots before accepting the results.) It has an accuracy of 99.8% or so. The first licensed test had an accuracy of approx. 93%. Tests under development range from 82-95.6%. These all use a system called ELISA (for Enzyme-linked Immuno- S? Assay) which is much easier to do than a Western Blot (Incidentally, Western Blot is so-called, because the first person to do "blotting" was E.M. Southern, so the blot was nick-named the "Southern Blot", and each new technique was given a different direction and the name stuck.) ELISA involves add sample, add reagant, check color change. Anyway, this new test has an accuracy of 99.6% (4 false positives in 1000 people) which is not much worse than Westerns. AND, of course the usual disclaimer: Although everyone with AIDS has been shown to have antibody to HTLV-3, the reverse is not true. The presence of antibody to HTLV-3 does not signify that a person has or will get AIDS, but has merely been exposed to it or a cross-reacting virus. -- Craig Werner !philabs!aecom!werner What do you expect? Watermelons are out of season!
sdyer@bbnccv.UUCP (Steve Dyer) (03/17/85)
> AND, of course the usual disclaimer: Although everyone with AIDS > has been shown to have antibody to HTLV-3, the reverse is not true. Um, I've heard that a significant percentage of AIDS patients DO NOT harbor HTLV-3 antibodies, and this bewildering state of affairs only complicates the application of such tests. Naturally, it isn't clear in these cases whether the absence of antibodies reflects absence of virus. Does anyone have the original references? -- /Steve Dyer {decvax,linus,ima,ihnp4}!bbncca!sdyer sdyer@bbnccv.ARPA