> >>Lynton:
> >>a possibility that what is being called a virus (HIV)
> >>could be "x": a new type of infectious agent
>
> >Carol:
> >They test for antibodies <snips> [which] can be found in the bloodstream
> >long after that infectious agent is gone.
>
> Lynton:
> That's interesting.  Is it usual to use antibody detection to diagnose other
> virus types as well, or do they have a more direct/specific methods?

Antibody detection is extremely popular these days.  I don't know, but
I wouldn't be at all surprised if it were the method of choice for well
over 90% of diagnostic blood tests.  Of course, there are many diseases
for which a blood test is not usually used (due to the presence of well
known signs and symptoms), even though one may be available.

> And another question: in cases where the immune system is severely disabled
> by medical treatment (eg chemotherapy for cancer) could AIDS-like symptoms
> develop?

Yes.  When AIDS first was defined, people in certain situations were
excluded from diagnosis with it because it was known that they were
already at high risk of immunodeficiency.  For instance, transplant
recipients -- they are given drugs which suppress their immune systems
so that there is less chance that their transplant will be rejected.
They have high incidences of all the diseases that AIDS patients are
famous for (except Kaposi's sarcoma, which appears to be linked more
to poppers than to any infectious agent).  PCP has been so common in
such patients that it used to be nicknamed "transplant pneumonia".
Please see Rethinking AIDS by Robert Root-Bernstein for more on this
topic.  There is far too much info on this for me to type it all here.
If you can't find a copy of that book, I could xerox a section or two
for you; just write me backchannel.

Carol