[Ip-health] WJS on $802: $282 million for clinical trials?

[Joseph DiMasi]

Jamie, I really don't wish to be rude and I am not being evasive, but I
do not have the time now to get into endless discussions on e-mail
lists.  So, this will be my last posting, at least until the study comes
out.  The study will have great detail on methodology, results, and a
discussion of issues.

You mention that Gardiner Harris reported that the methodology has been
criticized.  Anyone can criticize the methodology of any study if they
wish.  It doesn't make them right.  I would hope that people would
examine the methodology objectively and not let their political agendas
get in the way.  I know that this will not always happen.

You must know from the previous study and your own postings on this
topic, that a figure like the $282 million one is not an estimate of the
average clinical cost of a "single" drug.  It is an estimate of the
clinical cost per approved drug, inclusive of failures.

I find the numbers that you have on orphan tax credits to be irrelevant
to the question of what is the average resource cost of drug development
for the major research-based pharmaceutical firms.  They do not tell us
what was spent on total clinical period costs prior to approval even
just for drugs that received an orphan drug designation.  First, unless
it can be demonstratred that you cannot find enough patients in the
U.S., you do not get credits for foreign clinical testing.  Second, and
perhaps much more importantly, these drugs can be tested for non-orphan
indications, where the trial sizes and costs can be substantial.  Orphan
drug credits are not given for that testing, so those costs do not show
up in the IRS data.  Third, I assume from what you put out that you
included credits for all designations for which you could obtain data. A
good deal of those designations are for indications on drugs that have
already been approved for something else.  Testing costs on follow-on
indications are likely, on average, to be much lower than pre-original
approval testing because a good deal is already known about the drug at
that point.

Of course, orphan drug development is not at all representative of drug
development as a whole.  As you know, the trials are much smaller than
is the case for most other development.  So even if costs are relatively
low for testing orphan indications, as they likely are, they tell you
nothing about average costs for drug development as a whole.  It should
also be noted that our study was based on the R&D experiences of major
traditional pharmaceutical firms.  The vast majority of orphan drug
designations are given to small biotech and niche pharmaceutical firms. 
In terms of R&D budgets, orphan drug development is generally a very
small part of the development programs of the major firms.  For the
niche players it appears to be a low-cost low-return venture that they
are willing to undertake (with a handful of well-noted exceptions,
orphan drug sales are typically quite low).

As for other numbers that you cite with regard to clinical trial sizes
and costs per patient, they also are unconvincing evidence regarding
average full drug development costs.  Fast-track drug approvals are
likely atypical in that they would often have substantial post-approval
testing conducted on them.  You obviously don't pick that up in
pre-approval trial size numbers.  In any event, you can't impugn an
average cost estimate by going to the extremes of a distribution.  There
will always be some costs that are below average and some that are above
average.  If the only point that you are trying to make here is that
drug development costs vary, then fine.  Everyone involved in drug
development knows that, and I made the point that drug costs vary
substnatially in my presentation at our forum last Friday.

With regard to NIH trials, I don't know all of what they include in the
information you have obtained, but my understanding from people involved
in clinical trial testing is that, as good as these trials may be for
their intended purposes, they generally do not have the breadth and
depth of typical industry trials.  Thus, once again you likely have
something that is unrepresentative.  With regard to CRO information, as
far as I know there has been no published comprehensive study of actual
costs at CROs.  I don't know what is involved in the quotes to which you
refer.  I do know this.  I had obtained data from a company called
DataEdge several years ago on per patient costs.  They had information
on a large number of clinical trial protocols. For 1997 the average cost
figure was over $7000 with an upward trend.  However, their data were
"ONLY" for investigator fees and central laboratory costs. 
Pharmaceutical firms have large clinical trial infrastructures and have
to bear the costs of overhead, clinical trial monitoring, data capture
and analysis, interactions with the FDA and other regulatory
authorities, the cost of testing and producing clnical trial supplies,
etc.  Our study includes all these costs, although they are not broken
down into the various categories.  I should also note that our clinical
estimate is for the "clinical period" (as opposed to "clinical
trials").  The full estimate, in particular, includes the cost of animal
testing that occurs after clinical testing has begun.  My conversations
with industry managers indicates that the full costs may well be three
times what one would think of as the direct patient costs.  


Joe DiMasi



James Love wrote:
> 
> Joe, how do you get to $282 million for clinical trials?
> 
> As you know, while Gardiner Harris reporting that your study has been
> criticized on methodological grounds, I think the data itself will be
> hard to explain.   In particular, for clinical trials, something that
> one can check from observable data, in terms of numbers of patients
> cited by the FDA in trials, and the per patient costs of trials from the
> government and CROs, the "average" of $282 million isn't remotely close
> to what companies actually spend on trials.  This is about what the
> entire industry reported spending on their income tax for trials for
> 1997-1998 for all of the orphan's tested over this two year period, and
> here you report it as the average for a single drug!
> 
> The average number of patients cited by the FDA for HIV ARV drug
> approvals was about 1,200 patients, a significant number of them in
> government funded trials.  Are you saying the industry spend $235,000
> (risk adjusted, but before financing costs) per patient in these trials,
> after government grants?  Is this your idea of educating the public?
> How does this compare to the TB Alliance estimates of what the costs of
> trials will be?   How does it compare to the money the NIH spends per
> patient on cancer trials?
> 
>  Jamie
> 
> -------------------------------------------------------------------------
> 
> December 3, 2001
> 
> -------------------------------------------------------------------------
> 
> 
> Health
> Cost of Developing New Medicine Swelled
> To $802 Million, Research Study Reports
> By GARDINER HARRIS
> Staff Reporter of THE WALL STREET JOURNAL
> 
> PHILADELPHIA -- Researchers at Tufts University said Friday that the
> average cost of discovering and developing a new medicine has risen to
> $802 million -- a number bound to be swept up into the debate over drug
> prices.
> 
> The study said research costs have risen 2.5 times in inflation-adjusted
> terms since 1987, when Tufts found that the average research expense for
> a drug was $231 million. While total research costs increased 7.4%
> annually in the 1990s, clinical costs -- the component of research
> associated with testing drugs in humans -- rose 12%, said Joseph DiMasi,
> the study's lead author.
> 
> Merck & Co. Chairman and Chief Executive Raymond V. Gilmartin attended
> the unveiling of the Tufts data at a Philadelphia hotel and said
> increased clinical costs stem from demands by managed-care buyers that
> drug companies prove the value of their drugs in larger and longer
> trials.
> 
> "The number of patients in a new drug trial has increased from about
> 1,300 in the early 1980s to more than 4,000 for a typical new medicine
> today," Mr. Gilmartin said.
> 
> Dr. DiMasi has been issuing research-cost estimates for years based upon
> proprietary surveys of top drug companies. His numbers are routinely
> cited by industry backers to justify the ever-rising prices of new
> drugs. Just as often, consumer advocates dispute Dr. DiMasi's findings
> by criticizing his methods.
> 
> For instance, only $403 million of Dr. DiMasi's $802 million total are
> actual out-of-pocket expenses. The rest is an estimated cost of capital
> -- or the return that investing the money at an 11% rate of return would
> have earned over time. Bob Young, a researcher at the consumer group
> Public Citizen, says such accounting costs inflate the overall number.
> 
> "The bottom line is that pharmaceutical companies are still making
> wheelbarrows of money," Mr. Young said.
> 
> There are no winners in this dispute. Whether the average is $802
> million or half that, there is little doubt that the costs of drug
> research and the chances of failure in the lab are growing.
> 
> Big pharmaceuticals companies are spending much more on research in
> recent years, but their labs' output has actually fallen at the same
> time. Biotechnology companies have collected billions of dollars from
> investors, but very few have launched drugs.
> 
> "It's not necessarily disaster if what's coming out of the process is
> valuable and we as a society are willing to pay for it," Dr. DiMasi
> said.
> 
> China's Plan to Cap Drug Prices Upsets Makers
> 
> Pharmaceutical Firms File Suit to Block Michigan Effort to Cut Drug
> Prices
> 
> Merck Cuts Price for AIDS Drugs It Sells in China by About a Third
> 
> Bush Plan to Issue Medicare Discount Cards Suffers Temporary Setback in
> District Court (Sept. 7)
> 
> Mr. Gilmartin said his company is working hard to bring down the cost of
> research by investing in new genomics technology in hopes that better
> information will reduce the number of expensive failures in drug
> development. But drug-company executives have been touting for years the
> efficiencies that new technologies will bring to their labs, and each
> year the cost of research nonetheless goes up.
> 
> As for the politics surrounding the DiMasi study, Mr. Gilmartin said the
> DiMasi study sheds no light on drug prices. "The price of medicines is
> not determined by their research costs," Mr. Gilmartin said. "Instead,
> it is determined by their value in preventing and treating disease."
> 
> Mr. Gilmartin also argued that, given the enormous cost of research, big
> pharmaceuticals companies, not small biotechnology firms, are essential
> for developing medicines. He also said that patent-protection laws,
> which have come under attack by some drug-company critics, are vital to
> encouraging and protecting such huge investments.
> 
> The study found that the average development time for new medicines is
> 12 years.
> 
> Write to Gardiner Harris at gardiner.harris@wsj.com
> _______________________________________________
> Ip-health mailing list
> Ip-health@lists.essential.org
> http://lists.essential.org/mailman/listinfo/ip-health

-- 
**********************************************
Joseph A. DiMasi, Ph.D.
Director of Economic Analysis
Tufts Center for the Study of Drug Development
Tufts University
192 South Street, Suite 550
Boston, MA 02111
Ph: (617) 636-2116
Fx: (617) 636-2425
**********************************************