Back to the Top
The following message was posted to: PharmPK
Dear all,
I would like to hear your opinion on an issue that has kept my mind
busy for a while now. Let me provide some background first.
- We have develop a PK model for the autoinduction phenomenon of a
compound. The model should be applicable for induction, inhibition,
and displacement by other compounds, i.e., most PK drug-drug
interaction scenarios. In order to validate the model, I started to
look for publications, where drug-drug interactions have been
described, and where drug concentrations of both the inhibitor/inducer
and the affected drug have been measured. Fortunately, there are quite
a few recent publications, where the authors have done a good job in
detailing the interactions by measuring the concentrations of all
involved drugs and presenting non-compartmental analysis results.
Unfortunately for me, not a single author (out of 6 that I contacted)
would provide the underlying data to be used in my modeling project.
This despite the fact that the results were already published and I
offered to do it for free, all by myself or in collaboration with the
authors.
What I would like to have discussed is whether authors should be
required to make their data available to the public, once they publish
study results. The advantages I see with open data are use of already
available data in other ways than already tested, and possible check
of controversial results by looking at the source data. A second look
at the data may save resources to repeat studies, which should be
beneficial to everybody. Of course, I am not suggesting that the data
should be provided to the journal upon submission of a new article.
However, the authors should be willing to share the data used to make
published conclusions with other who might want to take a second look
at them.
There are of course several issues to be considered when requiring
open data. How would such requirement affect the choice of journal
(publish in journals that do not require this)? Is it practically
feasible to have such a requirement? Are there any legal issues? Why
would owners of data want or not want to make the underlying data of a
published work available to the public?
Any thoughts on this are welcome.
Toufigh
Toufigh Gordi, PhD
Clinical Pharmacology, PK/PD analysis consultant
www.tgordi.com
E-mail: tg.at.tgordi.com
Back to the Top
Toufigh:
You raised a most interesting issue. Have you had any answers?
The NIH has a clear requirement of the sharing of data obtained in NIH
funded research. See NOT-OD-03-032 and later notices.
We have a session on Ethical Issues at the Annual Meeting of the
American Association for Cancer Research, and the topic we discussed
this year was on ethical issues in publications. I am cc'ing one of
our key staff members at the AACR to see if she has any comments on
your question.
Back to the Top
Hi Walter,
I was aware of the NIH requirements and I believe they should be used
as a model, applicable to all research done in universities. I don't
remember now if the ones I contacted disclosed whether their work was
funded by NIH or not. However, a couple of them were done at
universities (US and Finland).
Thanks for your comments and interest!
Toufigh
Toufigh Gordi, PhD
Clinical Pharmacology, PK/PD analysis consultant
www.tgordi.com
E-mail: tg.aaa.tgordi.com
Back to the Top
Toufigh,
This is a problem I have some deep sympathy with. I cannot understand
the narrow viewpoint of people who wish to keep the details of their
published work a secret.
You may find the editorial written by my colleagues in Auckland on
this topic of some interest:
Anderson BJ, Merry AF. Data sharing for pharmacokinetic studies.
Paediatr Anaesth. 2009. (online)
Best wishes,
Nick
Back to the Top
The following message was posted to: PharmPK
Dr. Gordi:
You raise an important issue that needs to be addressed.
As you know, there are 2 types of science, the science of discovery
and the
science of integration. Your idea for open-source data will make the
later
easier than it is currently. It seems in the good old days, this was
easier
than it is now. An examination of many classic PK articles from the
70's and
80's will reval many narrative and intergrative reviews of PK topics,
often
containing multiple data sets. However in the good old days there was a
comparatively small cadree of PK scientists,and most of them knew each
other.
In todays world, because of globalization and the explosion of medical
research in "emerging economies" we are generating so much data that
time
and opportunity for integration is lacking. An open source data
requirement
may be the only way to achieve this. I would also suggest there is
much to
be learned from data that was subjected to less than optimal analysis
techniques. It is a shame that such data will be relegated to some
filing
cabinet (virtual or otherwise) to rot in obscurity, when it may hold the
secret to unlocking some pharmacokinetic mystery.
The science of informatics is progressing rapidly, and we need to take
advantage of developments in that field to help achieve the "open source
data repository".
Perhaps if we start small, by encouraging the development of "open data
repositories" at our respective institutions, we may be able to start a
ground swell.
Comments are welcome.
William R. Wolowich, Pharm.D., R.Ph.
Chair and Assistant Professor
Department of Pharmacy Practice
College of Pharmacy
Nova Southeastern University
Back to the Top
The following message was posted to: PharmPK
Let's see, this falls under intellectual property and companies have
paid
for it and you want them to share, possibly revealing strategies,
successes
and failures? And this information could save other companies from going
down the wrong path and wasting time and resources??? Do you have to
ask me that?
Back to the Top
The following message was posted to: PharmPK
Hi Ed,
Let me first make it clear that I am not suggesting anybody to share
data that is confidential and results of some analysis of it have not
been published. What I have difficulties understanding is why keep
data secret when the analysis results have been published for
everybody to see? Why would companies, or researchers elsewhere,
publish conclusions but refuse to share the underlying data? If a
company does not want others to know about their work, they have the
option of not publishing it, which is quite common. It doesn't make
sense to announce the results but withhold the underlying data. In my
belief, it is even unethical. Imagine this: in our next phase 2 study,
let's inform patients volunteering for the study that the data
collected from them will be used for analysis and interpretation by
our company on our own discretion. However, let's also inform them
that we will not share the data with others, who might have more
sophisticated tools and
knowledge of data analysis and who might find information hidden in
the data that we might miss. How do you think such statement will
affect your recruitment for the study? If I were a patient,
participating in a study to benefit me and others like me, I would
want to see the most being squeezed out of my data.
For those of you who want to read more about this, a very interesting
and relevant article by Drs. Anderson and Merry has been published in
Paediatr Anaesth. 2009 Jun 25 (doi:10.1111/j.1460-9592.2009.03051.x).
Toufigh Gordi, PhD
Clinical Pharmacology, PK/PD analysis consultant
www.tgordi.com
E-mail: tg.-at-.tgordi.com
Back to the Top
The following message was posted to: PharmPK
If we were all after the same thing-alleviation of pain and suffering-
it
would be important to share as much as possible. Unfortunately, profit
supersedes, limiting access.
Back to the Top
The following message was posted to: PharmPK
Toufigh:
I believe there is a growing body of scientists in a broad range of
disciplines that believe that sharing data is indeed an ethical
obligation (confidentiality issues aside). Here is an editorial from a
couple of years ago that states this pretty emphatically ....Nature
Biotechnology 25, 398 - 401 (2007) . There are multiple other
resources that make the same point.
Obviously the NIH has taken a strong stance on this issue. But I would
also argue that scientists from academic institutions need to examine
this issue very closely especially when they collaborate with private
industry. Refusing to share data may in some cases may place the
researcher in difficulty because it may violate the policies of the
academic institution (often times academic institutions take the
position that data generated on campus belongs to the institution not
the researcher and most certainly not to a private institution). Given
the notoriety of recent examples of academic misconduct, data sharing
may become an increasing requirement.
Data sharing by private institutions is another matter but more and
more journals are requiring more data be published in online
appendices for increased accountability.
Sincerely,
Carol Collins MD
Senior Research Scientist
University of Washington
Back to the Top
I think, primarily to prevent others (like Generics) from using that
data against the innovator companies in litigation suits. That is why
we always see critical data redacted in the FOI related to the IND/NDA
submissions.
Also, there could be a fear that researchers may find faults in the
manner the data was analyzed and how it was reported. But this could
be taken positively and be a part of the learning curve.
Manish Issar, Ph.D
Back to the Top
The following message was posted to: PharmPK
And if you as an employee of a pharmaceutical, etc, find an untoward
result, unexpectedly high tox from a novel drug, you will freely
publish this?
Back to the Top
The following message was posted to: PharmPK
Colleagues.
Isn't it true that secrecy and protectionism are the price our system
pays for innovation? Most people don't realize that the first reason
for the existence of patents is to get innovation into the public
domain, and protection of the inventor, be it individual or company,
is just a means to that end. It may appear to be the dominant effect,
and it may produce excessive profits, but in the longer term it does
get inventions into the overall market, as the generic market bears
witness.
My experience of working in several companies is that they would
welcome an individual who wanted the information for good,
collaborative reasons because they have a genuine interest, including
a commercial one, in finding the most reliable methods. They might
share information under quite strict confidentiality agreements and
that goes against the academic grain, but at least it's a halfway house.
Andrew Sutton
Back to the Top
The following message was posted to: PharmPK
Someone just wrote anonymously the following:
And if you as an employee of a pharmaceutical, etc, find an untoward
result, unexpectedly high tox from a novel drug, you will freely
publish this?
My answer would be that if I were such a company executive I would
seriously consider publishing that information freely if I thought it
would be in the public's best interest. However, the common scenario
is that the toxicity is produced by a novel molecule that nobody but
possible competitors would be interested in. You usually don't know
who else is on the same track molecular design-wise and all you do if
you publish is save the competition the toil, delay, expense and
disappointment that you have just suffered. It is quite a reasonable
approach and not always inflexible if some good reason to disclose the
information comes to light. That could happen for example if the
company is working with external consultants, as many do, and a
responsible company would arrive at a concensus on publishing that way.
Let me make it clear that I am not an apologist for bad behaviour in
the industry, far from it, but I do believe that we should stay as
rational and as well informed as possible. It seems to me that is the
way to derive good practice in the first place.
Andrew Sutton
Back to the Top
The following message was posted to: PharmPK
Observed toxicity is not necessarily the death knell for an
investigational
drug. There is no reason to report such negative data until additional
investigations have been done. The failure to address the observed
toxicity
with the additional investigations if the drug wasn't killed could be
the
death knell of the company itself. And rightfully so.
Mickey
Back to the Top
The following message was posted to: PharmPK
Here is some of what you are hoping for::
The AAPS BIOTEC Section is interested in forming a Therapeutic Protein
Immunogenicity Focus Group. The mission, vision, scope and goals of
the focus group are:
Mission:
To define the meaningful interpretation, presentation and clinical
application of immunogenicity, and to pursue validation of the risk
factors and underlying cause-effect relationships affecting the
immunogenicity of therapeutic proteins.
Vision:
Utilize cross-sectional expertise within AAPS to develop consensus and
innovative research solutions that influence drug development as it
relates to immunogenicity.
Scope:
Clinical immunogenicity of therapeutic proteins. Limited scope for non-
clinical (preclinical) immunogenicity when relevant to human
immunogenicity. Vaccines and non-protein immunogenicity are currently
out of scope.
Goals:
1. Develop recommendations and publish a white paper on sampling
considerations and the uniform reporting of data for appropriate
parameters such as incidence (per overall, and relevant categories
such as dose, immunomodulatory concomitant meds, demographics),
transience, persistence, early/late, NAb/Non-NAb nature, titer, etc
and PK/PD/safety correlations for clinical and product comparability
assessments.
2. Delineate the interaction of risk factors, immunogenicity and
tolerance mechanisms and possibility for development of biomarkers for
human immunogenicity and tolerance to therapeutic proteins through
research. Assess the feasibility of creating an innovative industry
sponsored competitive funding of academic research proposals
specifically designed to delineate tolerizing regimens that inhibit
ADA formation, and biomarkers of immunogenicity.
3. Design and execute an "AAPS Therapeutic Protein Immunogenicity
Consortium", comprising a database to collect and analyze anonymized
voluntary contributions of clinical data relating to cause-effect
attributes relationships such as immunogenicity incidence, NAb/Non-NAb
nature, titer, etc., against risk elements such as formulation
(impurities), manufacturing process variables, immunomodulatory
concomitant medications, molar-valence dose units, administration
route, dose frequency, duration of treatment, stability
characteristics (aggregates), and patient characteristics underlying
disease, genetic background, immune status) from across industry and
academia to lead to a meta-analysis that identifies the factors and
examines the nature and extent of their influence on clinical
immunogenicity.
Focus groups serve as incubators for programming ideas and create
networking opportunities with other scientists involved in similar
work. If you are interested in joining this new focus group, please
complete the form at
http://www.aapspharmaceutica.com/inside/Proposed_Groups/TherProteinFG/Proposed_TherProteinFG.asp
Please include your AAPS membership number and email contact
information with your communication. Once the focus group has been
approved by the AAPS Executive Council, a listserv will be created and
you will be notified of focus group activities and opportunities to
participate.
We are also looking for active volunteers to serve on the steering
committee. Participation on the steering committee includes monthly
conference calls, programming development, communication to our
members, and website updates. This is a great way to become involved
in AAPS.
Back to the Top
The following message was posted to: PharmPK
Andrew,
It seems I have difficulties understanding a simple, in my mind,
discrepancy. When detailed results of a study is published for
everybody, what is left to be kept secret? Again, I want to emphasize
that I totally understand, and agree with, companies who do studies
and want to keep the results for themselves, not publishing the
findings. This is part of the game and each study might provide
advantages compared to the competitors. What I don't understand is why
keep the data secret, while you tell the whole world what the outcome
of the study was?
In my simple example/experience, I have come across several drug-drug
interaction studies, where all the authors have done have been to
present changes in some non-compartmental analysis estimates of one
drug, mainly AUC, due to induction or inhibition of enzymes by another
drug. They already admit there is en effect. They announce how big the
effect has been and whether one should be careful using the two drugs
simultaneously or not. Now I want to go in and see whether the
induction or inhibition can be described another way, i.e. through a
compartmental model. If the original authors have done things
correctly, and if my proposed model is correct, we should arrive at
the same results. So why not share the data?
I must add that in a couple of cases the scientists publishing the
work were open for the idea of sharing data but their companies didn't
agree to do so. My guess is that in some cases, those deciding on
these matters are not scientists themselves and do not want to take
the "risk" of any unexpected event upon sharing data.
To re-phrase my earlier question to help me understand the whole issue
better, what would be in jeopardy for the "owner" of data, when
results based on the data have been made public? Why would one
describe the outcome but not how one got to those conclusions?
Toufigh
Toufigh Gordi, PhD
Clinical Pharmacology, PK/PD analysis consultant
www.tgordi.com
E-mail: tg.aaa.tgordi.com
Back to the Top
The following message was posted to: PharmPK
May I raise another ethical point.
Human subjects, particularly in later phase studies usually
participate on the basis that they are contributing to science and the
greater good. I have never seen a patient information sheet that says
the results will be commercial in confidence. Often the PIS will say
the information will be published and shared.
Matt
Dr Matt Doogue
Clinical Pharmacologist / Endocrinologist
Flinders Medical Centre, Flinders University
Back to the Top
The following message was posted to: PharmPK
Toufigh,
I think at the heart of this issue is the subjective nature of
pharmacokinetic data compared to other types of repository holdings
such as
proteomic or genomic data. As you are aware, just the choice of error
model
alone could change the findings, and indeed, there are numerous papers
with
pharmacokinetic data analysis therein for which I have felt the
results were
improperly derived. The implications of open access to data may lead to
open season on erratum notices from editors who have been contacted by
an
expert modeler, or not an expert; either way, someone can always have an
opinion that their model is better than someone else's. Other types
of data
are perhaps not subject to that same level of "art".
Realistically, no one will spend time as an exercise to reprocess
data, I
assume, however just the opportunity may threaten the financial
security of
those clinical pharmacologists and the like who have received grant
funding
based on (what was) a statistically significant observation.
Moreover, the issue of for-profit companies being expected or required
to
release data for published articles may reduce the number of peer-
reviewed
contributions that we depend on. If they have an obligation not to
"risk"
pending or future drug applications for the spirit of healthy scientific
competition outlined by Anderson and Merry, then maybe we would have to
expect the number to decrease. So in the end, any mandates by journal
publishers or editors that article acceptance and data release go
hand-in-hand, could have the opposite effect of what was initially a
good
intention.
That said, if any institution (preferably the CDER) decides to begin the
process, it should grow on its own without the need for ultimatums.
Don't you think?
Shawn D. Spencer, Ph.D., R.Ph.
Assistant Professor of Biopharmaceutics
Florida A&M College of Pharmacy
Tallahassee, FL 32307
shawn.spencer.at.famu.edu
|
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Access to data of published work" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)