[extropy-chat] magic johnson, aids, longevity ...

[Martin Striz]

On 4/17/06, ben <benboc at lineone.net> wrote:
> Martin Striz wrote:
>
> >> If antibiotic cocktails are not a good idea, why is this, exactly?
>
> > Because creating stronger selection pressures increases the rate of
> > evolution, in this case, of resistant strains.
>
> Um, then why does it work against HIV?

First, I want to make sure that you understand the distinction: we use
antibiotic cocktails against bacteria and antiretroviral cocktails
against viruses like HIV.

Second, drug cocktails work against bacterial infections as well, they
just aren't preferred for the reason I outlined.  So the question is,
why would you prefer them for HIV and not bacteria?  Because it's the
only thing that suppresses HIV to a significant degree.  Each of the
drugs alone does a poor job.  Obviously you don't want to accelerate
the evolution of any microbe, but preventing people from dying trumps
that.

Of course, our overuse of antibiotics has created multi-drug resistant
TB, which is immune to almost everything, so that's your trade off. 
As to why there aren't new immune strains of HIV yet, it's probably
because the cocktails we use attack multiple points in the viral life
cycle, so that the fitness path is too discontinuous to a totally
immune viral particle.  Most antibiotics act on the cell wall or the
ribosome, with a few exceptions.  It's easier to alter fewer
subsystems.  That's my guess, anyway.

> If this is right, then given the extremely fast rate of mutation in HIV,
> multi-drug therapy should fail. It should make HIV stronger. I don't
> think it's quite accurate to say that stronger selection pressures
> increase the rate of evolution. At the most, this would only be true up
> to a certain point. If the selection pressure became TOO strong, nothing
> would be able to adapt fast enough, and the organism would die off.
>
> So, for either bacteria or viruses, the way to avoid resistant strains
> arising would be to hit them hard enough, from enough different
> directions, that they had no chance of adapting.

That's my explanation above.

> It might be possible to devise a treatment strategy that traps the
> disease organism in an evolutionary local maximum, then changes the
> landscape so that the only way to survive is to leap to another peak,
> which is too far away for natural evolution to manage in a few steps.
>
> If we had 3 or 4 such treatments for any one organism, we could keep it
> at bay indefinitely, by switching treatments periodically. Any new
> strain would develop from the wild population, and have no evolutionary
> memory of the previously treated ones (cos they're all dead), so there
> would be no adaptation to this strategy over time.

That's probably the situation we have with HIV (crosses fingers).

Martin