<soc.support.depression.treatment>) wrote ...

<snip>

If I may use the above as an opening, I have a question for anyone who has
an answer:  In the beginning of the process, before any animal
experimentation is done, by what criteria do today's pharmaceutical
researchers choose which molecular entities (out of the infinite variety
possible) they will begin investigating in their search for antidepressant
candidates.  For example, do they set out by simply desiging/selecting a
series of molecules that appear likely to modify monoamine kinetics
(e.g. release/reuptake/receptor activity, etc. of those neurotransmitter
systems
popularly suspected of playing a role in antidepressant activity)?

Also, how frequent does it occur that molecular entities which have
successfully turned out to affect those neurotransmitters as so desired
(e.g. are SRI's, NRI's, SDRI's, XYZI's, etc...etc...) --and with
pharmacokinetic profiles akin to antidepressants in general (e.g. 'longish'
half-life, etc.)-- fail to exhibit any antidepressant properties?  4% of the
time?  20%?  50%?  90%?

The reason I'm posing these questions is to possibly gain some insight into
the significance of an antidepressant's classification by
neurotransmitter... (e.g. SRI, NRI, etc.).  In other words, _how much_ of
the antidepressant discovery process is based on the serendipity of
old fashioned "brute-force" trial-and-error, versus that gained by
researchers'
scientific knowledge/'hunches' of how antidepressants seem to work?  TIA

Sodah

This is a fairly complicated question, mostly because the way drug
research is done is fundamentally changing (or has fundamentally changed)
in the last couple of years.... but, traditionally the way it was done:

1) a particular chemical is obtained through trial and error hunches as
you suggest. These might be sourced from plants that have been identified
in various cultures, etc. like {*filter*} or mahuang.  Some {*filter*} were
generated because pharmacologists are crazy and they used to generate tons
of compounds and just take them and see what happens (LSD is an example).

2) the active agent is idenitified, purified, and often eventually
synthetically manufactured... like morphine or ephedrine

3) The drug is altered through a series of simple chemical changes. These
generally result in a change in its therapeutic properties. This is
generally the way {*filter*} get worse or better side effects, different
metabolism, {*filter*}-brain-barrier penetration, etc.  So codeine or
psuedoephedrine (or amphetamine).   A lot of times this is a really stupid
process that led to a lot of {*filter*}....

for example, a drug co wanted to generate a synthetic ephedrine... they
generated both amphetamine and pseudophedrine in this process. Both {*filter*}
are quite similar to ephedrine....

in the case of {*filter*}, the story is even stupider.  Basically, salicylic
acid was a drug used to treat pain, and Bayer discovered they could reduce
the side effects by adding acetyl groups (i.e., aspirin = acetylsalicylic
acid).... so they tried this with morphine, and generated {*filter*}
(diacetylmorphine). it was even marketed as "just like morphine without
all the {*filter*} side effects... from the makers of Aspirin." of course,
they really didn't know what they were doing, which is often the case.

4) In the process of doing step #4, a lot of times the drug will come to
have an entirely new mechanism of action, and a new drug class is born
from it.

- as far as the number of compounds generated and tested, these are in the
hundreds, they tend to have just the manufacturer's identifier, followed
by the number of the compound. e.g., RU-486, Mk-801, etc.  Pretty much
everything is derived from something else.

Now all that is changing. Whereas traditional pharmacology has focused on
receptors on the cell membrane or steroid receptors, a radical shift
toward molecular (read: genetics) pharmacology has occured in the last few
(very few) years. Most research in {*filter*} right now is focused on altering
gene expression.  In this case, a specific gene product that is deficient
or overexpressed has been identified for a particular disorder and
recently tools have been developed which can alter this expression (or
even knock it out entirely).  The shift toward molecular pharmacology is
radical but has been fairly complete, pretty mcuh everyone has jumped on
the bandwagon, because the power that {*filter*} can have  in this domain is
much greater.

Cheers,
Stephan