Subject: Re: Examples needed against Soft Patents
From: "Karsten M. Self" <>
Date: Sat, 1 Jan 2005 14:26:54 -0800
Sat, 1 Jan 2005 14:26:54 -0800
on Mon, Dec 27, 2004 at 01:22:42PM +0900, Stephen J. Turnbull ( wrote:
> >>>>> "kms" == Karsten M Self <> writes:
>     kms> Open peer review != banning pharmaceutical patents.
> Did I say that?  Open peer review (ie, publication of the formula at
> patent application) will shorten the useful life of the patent
> appreciably by giving a leg up to those trying to develop a competing
> product.


Are there any highly successful drugs that *don't* already have
signficant generic competition when patent ages out?  Would opening peer
review change the dynamics markedly?  I could see possibly reduced
costs, but suspect that the impacts would be pretty muted.  The major
costs in drug development are:

 1. Initial compound development:  researching and evaluating the
    hundreds of candidate compounds which might result in a single
    marketable product.

 2. Phase II/III trials.  FDA review process for efficacy and safety.

When you're doing generic product, you've eliiminated 1 & 2, *and* the
current system already supports this.

Incidentally, in discussing open methods and pharmaceutical development,
I've just run across a few mentions of Steven Weber's  The Success of
Open Source , which apparently extends the discussion to the
pharmaceutical industry.  Weber's on the PoliSci faculty at Berkeley:,S/
>     kms> It does imply that testing or testing data be available to
>     kms> multiple entities, preferably with at least independent, if
>     kms> not non, biases.
> What makes you think any such multiple testing will be done?  

Well, I know that if the data *aren't* available, it  won't  be done.
That's a given.

There are groups with an interest in this, ranging from competitors to
medical researchers to public interest groups.

> You know very well how expensive it is, and that one of the complaints
> of the Abolish the FDA crowd is that the  required  testing is already
> overkill, which prevents useful drugs from getting to market in time
> (in some cases, at all)?  

There's a significant difference between initial FDA reviews and
followups.  The FDA process is highly formalized and intensive.
Clinically and statistically valid follow-on studies would likely be
feasible at far lower cost.

> What makes you think that making the data available is going to be
> terribly useful (given the atrocious example of the way patent claims
> are written)?

This flips our CALs discussion around.  I'm discussing general theory
(publically available data), not a specific implemenation (patent
application data and/or claims support).  I'll leave this one out of the
discussion at present, but note that existing patent claims authoring
requirements are likely not sufficient.

>     kms> There's the side issue of much basic pharmaceutical research
>     kms> (an area with which I have a passing familiarity) being done
>     kms> either at publicly-funded research institutions, or by
>     kms> independent development labs, these later being hired or
>     kms> bought outright by the major Pharma players.
> So?  I don't see your point.  Those independents have strong interest
> in keeping their discoveries to themselves in order to maximize the
> first-copy price (according to the famous paper by Boldrin and Levine,
> and Quah's followups, they can extract the full monopoly rent in one
> shot by doing so, even with no patent protection at all---of course I
> don't believe that result can hold in practice, but it shows what the
> incentives are).  

There's monopoly interests, on the one hand.  There's the fact that the
research is publicly funded, and according to some, should have a public
interest component, other than monopoly market development, as a result.

> So what does it matter, as long as they must depend eventually on the
> resources of a big player to do the development (ie, testing)?
>     kms> Pharma itself is largely in the business of marketing
>     kms> existing products (not an inconsiderable or deprecated
>     kms> activity of itself), rather less so independent research.
>     kms> I've been involved in several projects whose major objective
>     kms> was extending an existing drug's indications rather than
>     kms> safety & efficacy of new compounds.
> Ie, efficient innovation (improving service at minimum cost).  Sounds
> like a great idea to me!  Again, I don't see your point.

It's a question of research allocation, market efficiencies, social
good, and private/public benefit.

Taking a case from recent history, there's a shortage (in a clinical if
not an "efficient markets" sense) of vaccine and/or treatements for
epidemic conditions or threats including influenza, malaria, smallpox,
and AIDs.  This is largely driven by the market circumstances of
treatment:  a single dose is required, of a large number of patients, on
an occasional (annual or less frequent) basis.  Or the risk is
theoretical, hard to predict, small incedence, large effect[1].  Or the
vulnerable populations are largely without economic means for payment.  

The area of antibiotic treatments is aother that's considered
underserved by the current Pharam industry.

The economic rewards come from low-risk situations:  take an existing
product (no research costs), look for a chronic condition (obesity,
heart disease, diabetes, depression) associated with a wealthy (and
insured) population.  Wonderful business case.  From a public health
perspective, somewhat less desireable.



1.  Somewhat like getting hit by an asteroid:  odds of occurance in any
    one year are low, but if/when it happens, the effects are large.

    Tsunamis might also come to mind.   Care to estimate the cost per
    life saved of the US Pacific tsunami detection network?  Big waves
    in the 50s and 60s.  Hilo lost 150 in the South American quake.
    $200,000 buoys and a satellite data network.  The surfers love it
    for its predictions, but can this really be justified on the basis
    of lives save? 

    Mind that my own view is:  yes.  The time period is sufficiently
    short, and incidents so severe, that in the long run, after we're
    all dead, it will pay off.

Karsten M. Self <>
 What Part of "Gestalt" don't you understand?
    I've got my country's 500th anniversary to plan, my wedding to arrange,
    my wife to murder, and Gilder to frame for it. I'm swamped!
    - Princess Bride

["application/pgp-signature" not shown]