The Gene Patenting Issue:
Urgent Problems in Science Policy and Law

Presented to the Panel on Gene Patenting of
the Office of Technology Assessment, March 25, 1993

David J. Galas

In many ways, the Human Genome Project (HGP) epitomizes the promise of the biological
sciences for the future. The impact on our lives of the vast amount of new knowledge of the
living world is just beginning to be realized - it will inevitably transform medicine, agriculture,
food manufacture, chemical technology, and many other industries in ways that are difficult to
predict. The HGP is also a project that holds within its promise the capacity to raise some
significant and difficult issues concerning human genetic information under the law. Among
these are the issues you are beginning to grapple with today - questions related to the conditions
and constraints under which human genetic information can be patented, the implications for the
future of scientific inquiry, the commercialization of scientific advances and of other uses of
genetic information by society.

The way to approach the complex of issues most effectively, I think, is to consider what the most
desireable outcomes for science and the applications for society are, and then to devise and
consider appropriate policies and guidlines that can achieve those outcomes. I wish to leave you
with the realization that there are fundamental issues involved and that solutions to these difficult
problems are needed urgently.

The Department of Energy (DOE) has sponsored the Human Genome Project (HGP) from its
initiation, and DOE and NIH now jointly support this research effort. We consider this historic
project of immense potential value to the American people and to the world. We strongly
support the expeditious transfer of the benefits of this research and of the associated technologies
to the public. For DOE and for 11 other Federal agencies, there exists a real imperative for the
rapid and effective public use of the results of biotechnology research, including the HGP and
many other research efforts.

In the HGP the overall approach to elucidating the genetic contents of the genome is first to
physically “map” it and then sequence large parts of it, in order to create a knowledge base of
unprecedented detail and complexity, so that subsequent research can be more effective and
efficient. This process is well underway and is making spectacular progress. Never before has
so much genetic information been gained, so many genes been located, identified or
characterized. Never before has the technical means to gain information been more promising.
No matter how strongly I say it here you will undoubtedly be surprised in the next few years by
the sheer rate of information acquisition. This prospect makes it critical that we squarely face
the problems of how to deal with the patenting of any genes - but particularly of human genes.
Though the problem has been with us for some time, it has never been adequately dealt with.

The recent controversy engendered by the patent applications based on partial cDNA sequences
submitted by the NIH has actually provided an important impetus to finally address this problem.
It is an impetus we should heed well, and use in order to grapple with the immediate issues, but
the broader issues must also be dealt with. I will argue here that the NIH patent applications
themselves are certainly not the most critical issues facing us.

I wish to say at the outset that I strongly support the aggressive pursuit of a resolution to the
immediate issues raised by these applications. This is clearly important for both the commercial
sector and the international scientific research community. While we were not supportive of the
point of view represented by the original filing by NIH, we are certainly not against patenting,
nor against patenting genes. On the contrary, we are in favor of a wide application of the patent
mechanism to protect rights to inventions, and to open up the transfer of knowledge and
inventions for the public good. The full mechanism must also include various licensing
arrangements, however, which also deserve full and informed consideration.

In a general sense, we of course recognize and support applications of the patent process that
fulfill the criteria laid down by the Constitution of the United States that they "promote science
and the useful arts". In the specifics, however, lies the difficulty. These are time-honored
criteria that focus our attention on some of the most important questions about biotechnology
patents: do they promote or inhibit the vigor of science; and, do they promote the development
and commercialization of biotechnology; e.g., "the useful arts"?

2
Turning to the specfic case of the now notorious NIH cDNA patent applications, consider some
of their potentially unsettling effects. These include:

a) The sequences of anonymous cDNA fragments, of no known function, cannot reasonably be
described as “complete or discreet inventions" - rather they are intermediate research results.
For technical reasons, including the incomplete nauture of the sequences, we expect that patent
applications like this will likely lead to a tangle of disputes requiring judicial resolution, and
delay and dilute the benefits of the research.

c) If acDNA patent were to be issued (this seems unlikely but by no means foreclosed at the
moment), such patents could serve as a basis for claims against later, truly creative and novel
inventions that have a clear and specific use, and that are developed independently of the partial
sequence. A cDNA patent based on an anonymous fragment could in this way be a strong
disincentive to investment and economic exploitation of the results of the HGP.

c) The potential for inhibition of free collaboration and rapid exchange of data, both nationally
and internationally, is real. This is not an evident, or necessarily expected effect, but this
inhibition has been observed. Inhibition of free exchange would reduce the effectiveness of our
research in many areas.

We are also concerned that the logic of the NIH patent applications, if it were applied rigorously
as it is, could put all genomic sequences, mapping information, DNA fragments, clones (YAC’s,
cosmids etc..) into the same category as the cDNA fragments. All "intermediate research
results" could serve as the basis for broad, relatively unspecific claims on the fundamental
structure of the human genome. In my view, it is this kind of broad, sweeping claim that is
clearly the dominant problem, and not the patent applications on anonymous cDNA patent
applications themselves.

All this being said, the NIH patent applications really represent only a clarion call, a warning
that the system may not be equal to the strains placed on it by modern gene technology. They
represent only the beginning of our problems. The most important issues, I would argue, do not
concern patentability, but rather concern what claims on genetic sequences can be protected by
the patent system, and what constraints on use and availability to society are desireable.

Knowledge of Nature versus Invention:
A central question for your consideration is one that extends well beyond the cDNA patenting

3
case: where should the line be drawn between entirely appropriate patent protection for useful
inventions and unwarranted claims on the ownership of fundamental knowledge of nature? What
distinguishes the basic genetic information that is the fundamental currency of biological science
from, for example, the spectrum of neon or any other element that is the basic currency of
atomic physics. This question must be answered with due consideration for the good of private
industry, the long-term vigor of the scientific enterprise, the benefits to the American public, and
the benefit of the international community. Other ways of asking the question might be: where
is the boundary between fundamental knowledge about ourselves and those useful inventions that
patent law properly protects; where should we draw that line between basic research results and
a product or process? Central to this issue, in turn, is the closely related question of what rights
patents on genes should really protect. This important aspect is touched on in the next section.
These are not really new questions, but there are new features here specifically related to genetic
information, and there is a new urgency.

Use versus Ownership:

Is genetic information qualitatively different from other natural information? Should our own
genes be subject to the same property values implicit in our current “intellectual property" laws?
What does it mean for an individual or a company (or a Government agency) to "own" a gene?
Given that we can define the rights and claims entailed, should "ownership" rather than a
particular utility be the focus of patent rights? These are among the difficult questions that now
must be addressed. Given that the function of a specific gene in question is known, the most
significant intellectual property question is probably: should a patent on that gene include the
right to all possible future uses of that information, the broadest kind of claim, or should the
patent claims should be confined to the utilities specified, a quite limited kind of patent. The
distinction is fundamental, I think - ownership or uses, broad far-reaching claims or limited well-
specified claims.

While practice in the US has been strongly in favor of "ownership", I propose to you that careful
consideration must be given to strengthening the important notion, already inherent to a limited
extent in US law, of only allowing the patenting of DNA sequence information for specific,
anticipated uses, and only for those uses - the only difficulty in this, other than that inherent to
a shift of practice, would seem to be in clarifing how specifically the "use" needs to be defined.
Though there are, I understand, recognized practical problems with “use patents" versus
"structure of matter patents", a sufficient emphasis on utility in patents on genes might deftly
accomplish at once several important goals. It could actually resolve the present NIH patent

4
controversy, guide future policy on problems like that discussed below, and refocus our attention
on the real purposes of the patent law.

Patenting the T-cell Receptor Genes?

Now I would like to describe a specific situation that has recently arisen which epitomizes the
most difficult questions above, and suggest a possible course of action. This example - one of
several possible - could provide a concrete, practical and immediate issue on which to focus an
otherwise broad and relatively abstract deliberation. A research group supported by the DOE
Human Genome Program has just completed the DNA sequence of the genes determining a major
component of the human immune system - the T-cell receptors. About 450,000 base-pairs of
sequence of a major site of this information in the human genome (the beta locus) have been
obtained. The question is: of the primary sequence information, the known and unknown
functions encoded in this information, and potential utilities, what is it that can and should be
protected by patents?

Something of the nature of the genetic information in the T-cell receptor genes should be
understood before the issue is discussed further. The human immune system depends on a set
of cells called T-cells as a fundamental part of the apparatus for the immune response. These
cells have proteins on their surfaces that determine the specificity of their immune activity. They
have the same enormous repertoire of binding specificities that circulating antibodies have, which
is produced by a combinatorial mechanism similar to that used to produce the enormous variety
of antibodies. It is illustrative of the importance of the T-cells that it is the destruction of one
class of these cells that is the central, lethal effect of the AIDS virus. The T-cell surface
proteins, called T-cell receptors, are encoded by several complexes of gene segments located
together in the T-cell receptor loci - one of which (the beta locus on chromosome 7) has now
been sequenced.

These gene complexes encode key components of the human immune system, and are also, by
the same token, implicated as the locations of genetic alterations that may cause, or contribute
to, a variety of autoimmune diseases, or suceptabilities to diseases. So far only a genetic marker
for multiple sclerosis has been clearly identified in the T-cell receptor locus, but it is likely that
markers for rheumatoid arthritis, and a variety of other autoimmune diseases, will also be found
in this region. This region is clearly one that holds the key to a number of prevalent, serious
human diseases.
Because the function of the T-cell receptors are known the uses that could be described for a
patent application on this DNA sequence are many, both for identifying the important sequence
variations that are linked to disease states, and for possible therapeutic purposes. The utility
would therefore be very much more compelling than that described for the cDNA’s in the NIH
patent applications. Since the uses and applications associated with the immune loci are much
more identifiable and more closely related to the sequence data, the problem is much more acute,
concrete and perhaps more difficult. In addition, since the T-cells are a major component of the
immune system, anything involving cellular immunity could conceiveably be covered by the
broadest of claims.

While I am inclined to think that the genetic structure of the human T-cell receptor loci itself,
like most of the human genome, should be viewed as fundamental knowledge of nature, and not
as an invention, the issue is complex and difficult and present practice does not seem to make
the distinction well. The line has never been definitively drawn between these two types of
knowledge, especially in the biotechnology areas. The question is even more poignant in this
case in which we are dealing with the genetic structure of the human body, and with genes
closely linked to human diseases. The question of "ownership" versus "uses" apears to be a
sharp one here. In the face of these difficulties I would like to propose a specific course of
action for your consideration.

The researcher leading the T-cell receptor sequencing project, Prof. Leroy Hood, intends, with
our full support, to file, within the next few months, a patent application on the DNA sequence
of the human T-cell receptor beta locus for the specific uses of developing the diagnostic and
theraputic tools for dealing with specific autoimmune diseases, as outlined above. At the time
of the filing it is intended that all the sequence information will be submitted to GenBank, the
sequence database, and thereby released for the scrutiny and use of the international scientific
community. In our view, this action - patenting in the strict context of uses to be made of the
information, with claims covering only specified utilities rather than broad “ownship" claims -
should have several important effects. It will make this important sequence information
immediately available to the scientific community for the study of the immune system. It should
protect specific rights for the investigator but not block future rights to important, novel
inventions yet to be made using this information, and it should stimulate the consideration of
some important questions on gene patents.

What is being proposed here would seem to be in sharp contrast to past practice. The recent
example of the patenting of the erythropoieten (EPO) gene is illustrative of this contrast. When
the structure of the EPO gene and the function of the gene product in stimulating differentiation

6
of red blood cells was determined, the patent claimed the structure of the gene and all its possible
future uses. No matter what might later be discovered about this gene, or invented using it, the
patent would establish prior rights to those inventions - virtual "ownership" of the gene. What
is intended for the T-cell receptor genes is that only uses directly related to what is now known
and can be specified will be claimed. In this sense, “ownership " and "uses" are contrasted.

A very recent discovery can serve to emphasize this point further. The gene in which a mutation
leads to "Lou Gehrig’s disease” (ALS) has just been identified. It happens that the gene product,
superoxide dismutase, has been known for many years - it is an enzyme that supresses free
radicals in the cell. The gene has also been known for some time and has been mapped on
chromosome 21 for more than 2 years. If the gene had been patented, and therefore "owned"
. in some sense, there could be no new claims on it by the disocverers for diagnostic or therapeutic
purposes related to ALS. If, on the other hand, the gene had been patented only for the uses that
could be specified before the recent discovery, the way would be clear for the newly discovered
role of the gene to lead to valid patent claims for this unanticipated utility. This latter scenario
would seem to be much more in the proper spirit of the patent system.

It is vital that the research use of sequence data and the rights to future inventions using this
information be protected. I strongly encourage you to look closely and wisely at the above
example of the T-cell receptors, and consider whether the suggested course of action will be
useful, and also whether the patent system needs to be changed to permit this to work well. It
is Clear that this example is an early indicator of an abundance of rapidly approaching problems
stemming from the imminent deluge of genetic and sequence information.

Conclusion:

In the final analysis, the vigor and competitiveness of a country’s technological economy is
driven by the vitality of its scientific research enterprise. The vitality of the US research
enterprise has been remarkable and has yielded benefits in a wide spectrum of scientific
disciplines. That vigor depends critically on the continued free exchange of information about
scientific research data, results, methods, and tools. The quality of both science and application
will be negatively affected by anything that inhibits, or even slows, the free exchange of this —
knowledge.

The purpose of Article 1, Section 8 of the Constitution, "...to promote the Progress of Science
and useful Arts," is vague, and has been interpreted in many different ways. Ultimately, it is

7
the hard job of Government to take the long view and try to see what is best for the future of
the scientific enterprise itself and for the vigor of science and technology-based industry. It is
also the job of government to bring some clarity to the present chaotic situation with respect to
gene patenting. The clear benefit to both of a solid and fair patent system is particularly
important to a thriving biotechnology sector. Biotechnology has now presented us with some
serious problems that extend the discussion of gene patents to questions of the boundary between
fundamental knowledge of nature, including ourselves, and useful inventions, and the question
of very nature of property rights for genes. These questions must be addressed. I have
presented you with an outline of the central issues and an example that epitomizes the nature of
some of the difficulties. I trust that you will wisely consider its implications. The thoughtful
deliberations and recommendations of this panel should provide a catalyst for careful
formulations of future policy, national and international, that will lead to a stronger and wiser
system.

oie is 3