US Treatment of Reach-Through Claims and Reach-Through Royalties
Frank P. Grassler
Genomic patenting poses many questions and problems, one of them being the issues raised by patent claims in the specialized area of reach-through claims, and another one being the issues raised by whether an infringed patent is entitled to an award based on reach-through royalties on sales of products invented by others. This paper will discuss these two distinct subjects, beginning with the subject of reach-through claims.
1
Reach-Through
Claims
1.00 Reach-Through Claims in
General
Reach-through claims (sometimes loosely referred to as reach-through patents) are at once a special subset of both genomic patents and research tool patents. Inventors, investors and managers typically take the view that they want to get reach-through claims on their inventions, particularly when a nucleotide sequence is potentially useful in drug discovery technologies. The goal would be to have a claim that takes a novel nucleotide sequence and have that claim ultimately cover or read on an as-yet undiscovered drug that to some extent came about as a result of using the expression product of the nucleotide sequence.
The term “reach-through claim” however, is actually almost a pejorative term, since it defines claims that tend to have two fundamental and important flaws. These flaws are inherent in this claiming style since the object of such a claim is to reach through the boundary, or metes and bounds, of the claim to cover products that will be invented by someone else, even though it is well settled that the claim must have understandable bounds so that the public can know with certainty what is part of the invention and what is not. To put it another way, the claim, to be valid, must not reach through (beyond both literal infringement analysis and doctrine of equivalents analysis) to that which is not part of the present invention and is invented by another. This is not to say that a broad claim to a pioneering art cannot read on improvements or that generic claims cannot read on undisclosed embodiments-clearly it has always been the case that they can. But where the fundamental requirements of utility, enablement, description and to a lesser extent obviousness are not met in a way that results in the claim extending its coverage through its legitimate bounds onto to the inventions of others, the claim is not permissible and is said to reach through.
We will first set out some illustrative claim types that are typically seen in connection with reach through problems. We will then examine the requirements in the US of description and enablement below, but first review the requirements for utility, since that is a threshold to further examination of a claim.
1.01 Representative Claim Types
1.02
The
Receptor/Small Molecule per se Claim
Here the
molecule is defined as binding to a target (which may or may not be a receptor,
but can include other classes such as enzymes) but the molecule is not yet
specifically identified (for example “A receptor X agonist”, where X is a
defined biological receptor protein). Here, although the receptor has been
identified, the complete list of those ligands (we will use the terms ligand,
agonist, antagonist and compound interchangeably here) that bind to the
identified receptor is not comprehensively listed in the specification. The
specification may disclose one, several, or a relatively great number of
ligands, but not all those that the inventor regards as part of the invention.
In general, the specifications of such applications reveal little real drug
discovery, especially when contrasted with the scope of drug discovery efforts
that go on in chemistry-based vertically integrated pharmaceutical companies.
The specifications typically mention compound classes in a generic way e.g.
triptans, imidazoles, etc., but provide relatively little or no information
regarding specific compounds that actually are shown to modulate the receptor
in question or that had a reasonable expectation of success in modulating the
receptor if one were to follow the teaching of the specification. As anyone who
has had experience in prosecuting organic medicinal patent applications knows,
a sufficient number of compounds in any given class has to be given that will
demonstrate the importance of a core structure or scaffold in predictably
modulating the receptor in question.
Additionally,
one frequently sees claims similar to this in structure, except that what is
being claimed is a monoclonal antibody that will recognize and bind to the
defined receptor.
1.03
The Method
of Screening Claim
Here the molecule is defined as having been shown to interact positively with a defined receptor in an analytical screening arrangement (for example, “A method of identifying an agonist of a defined target comprising contacting a candidate agonist with said target under defined experimental conditions”). Here again, although the receptor has been identified and an experimental or analytical apparatus or set-up has been disclosed, the complete list of those claimed ligands that bind to the identified receptor is not comprehensively listed in the specification. A typical claim that depends from this type of method claim is to any ligand identified by this method.
1.04
The
Functional Use Claim
The claim here is to a method of treating a disease by a compound defined not by its structure or structural formula but rather by (a) its ability to bind to a target (for example “A method of treating disease [Y] by administering a compound which is a receptor [X] agonist.”) or (b) by its ability to bring about a biological response of interest (“A method of inducing plasma levels of a secreted protein by administering a compound which is a receptor [X] agonist.”) or even (c) by virtue of having been identified by the method claim as illustrated above. These are somewhat hybrids of the first type of reach-through claim and suffer from the same deficiencies.
2
35 U.S.C.
§101 and §112, ¶ 1-Utility
2.00
US Patent
Office Definitions of Specific, Substantial and Credible Utility
The United States Patent and Trademark Office (USPTO) has issued
guidelines that set forth the U.S. patentability requirements of “utility” and
“written description”[1] The
Guidelines represent the USPTO’s current understanding of the statutory
requirements of utility and written description.
2.01
Credible Utility
Where an applicant has specifically asserted that an invention has a
particular utility, that assertion cannot simply be dismissed by a USPTO
Examiner as being “wrong”. Rather, the Examiner must determine if the assertion
of utility is credible (i.e., whether the assertion of utility is believable to
a person of ordinary skill in the art based on the totality of evidence and
reasoning provided in the application). An assertion is deemed to be credible unless
(A) the logic underlying the assertion is seriously flawed, or (B) the facts
upon which the assertion is based are inconsistent with the logic underlying
the assertion. Credibility as used in this context refers to the reliability of
the statement based on the logic and facts that are offered by the applicant to
support the assertion of utility. A credible
utility is assessed from the standpoint of whether a person of ordinary
skill in the art would accept that the recited or disclosed invention is currently
available for such use. For example, no perpetual motion machines would be
considered to be currently available. However, nucleic acids could be used as
probes, chromosome markers, or forensic or diagnostic markers. Therefore, the
credibility of such an assertion would not be questioned, although such a use
might fail the specific and substantial tests that are explained
further below.
2.02
Specific Utility
A utility next has to be specific to the subject matter claimed.
This contrasts with a general utility
that would be applicable to the broad class of the invention. For example, a
claim to a polynucleotide whose use is disclosed simply as a “gene probe” or
“chromosome marker” would not be considered to be specific in the absence of a disclosure of a specific DNA target.
Similarly, a general statement of diagnostic utility, such as diagnosing an
unspecified disease, would ordinarily be insufficient absent a disclosure of
what condition can be diagnosed.
A utility must define a “real world” use. Utilities that require or constitute carrying out further research to identify or reasonably confirm a “real world” context of use are not substantial utilities. For example, both a therapeutic method of treating a known or newly discovered disease and an assay method for identifying compounds that themselves have a “substantial utility” define a “real world” context of use. An assay that measures the presence of a material which has a stated correlation to a predisposition to the onset of a particular disease condition would also define a “real world” context of use in identifying potential candidates for preventive measures or further monitoring. Conversely, here are some examples of situations that require or constitute carrying out further research to identify or reasonably confirm a “real world” context of use and, therefore, do not define “substantial utilities”.
A.
Basic research such as studying the properties of the claimed product itself or
the mechanisms in which the material is involved.
B. A
method of treating an unspecified disease or condition. (Note, this is in
contrast to the general rule that treatments of specific diseases or conditions
meet the criteria of 35 U.S.C. § 101.)
C. A
method of assaying for or identifying a material that itself has no “specific
and/or substantial utility”.
D. A method of making a material that itself has
no specific, substantial and credible utility.
E. A
claim to an intermediate product for use in making a final product that has no
specific, substantial and credible utility.
Note that “throw away” utilities do not
meet the tests for a specific or substantial utility. For example, using
transgenic mice as food for animals that feed on mice is a utility that is
neither specific (all mice could function as such food) nor credible (spending
thousands of dollars to genetically engineer a transgenic mouse and then using
it as food is not a “real world” context of use; foolhardiness, though common
enough in the world, is not “real world”). Similarly, use of any protein as an
animal food supplement or a shampoo ingredient are “throw away” utilities that
would not pass muster as specific or substantial utilities under 35 U.S.C.
§101. This analysis should, of course, be tempered by consideration of the
context and nature of the invention. For example, if a transgenic mouse was
generated with the specific provision of an enhanced nutrient profile, and
disclosed for use as an animal food, then the test for specific and substantial
asserted utility would be considered
to be met.
2.04 Well
established Utility
There may exist a specific, substantial, and credible utility which is
well known, immediately apparent, or implied by the specification’s disclosure
of the properties of a material, alone or taken with the knowledge of one
skilled in the art. "Well established utility" does not encompass any
"throw away" utility that one can dream up for an invention or a
nonspecific utility that would apply to virtually every member of a general
class of materials, such as proteins or DNA. If this were the case, any product
or apparatus, including perpetual motion machines, would have a "well
established utility" as landfill, an amusement device, a toy, or a paper
weight; any carbon containing molecule would have a "well established
utility" as a fuel since it can be burned; and any protein would have well
established utility as a protein supplement for animal food. This is not the
intention of the statute.
2.05 Utility and US Courts
One must
bear in mind that the Court of Appeals for the Federal Circuit (CAFC), which is
under no obligation to follow the PTO Guidelines, may not ultimately agree with
the restrictive scope of the PTO’s recent Utility Guidelines. Although the CAFC
supports the USPTO’s interpretation of the relationship between §101 and §112[2],
the CAFC has on the other hand tended to
consider utility a very low threshold barrrier to patentability. See, for
example the 1994 case of In re Alappat, [3] where an algorithm that merely produced smooth
waveforms was found to have the requisite usefulness. The CAFC has gone so far as to say that a
claimed invention must be totally inoperative to lack credible utility[4],
and declared a willingness to accept in
vitro evidence or data as establishing, for example pharmacological utility
in the absence of clinical studies[5].
In drafting
patent applications, in the field of biotechnology where a reach- through claim
is most likely to be found, applicants should take note of the relatively
favorable decision of the CAFC and take care to formulate some sort of
practical utility within the specification, and if possible to tie that
practical utility to plausible mechanisms of actions since after all:
“Obviously, if a claimed invention does not have utility, the specification
cannot enable one to use it.”[6]
3
35 U.S.C.
§112, 1st Paragraph, Written Description
3.01 Description in General
The first key
flaw that makes a claim impermissibly reach through is that the specification
does not reasonably convey to the person of ordinary skill in the art a
description of the completed invention, in the sense that the person of
ordinary skill is not left convinced that the applicant was actually in
possession of the claimed invention as of the date of filing the application.
In general, the rule is that if the specification conveys with reasonable clarity that the inventor was in possession of the invention as claimed as of the effective priority date, then there is adequate written description.[7] The criteria by which a claim will be examined in the USPTO are based on a line of cases, the most important of which is The Regents of the University of California v. Eli Lilly & Co.[8] This case, which was decided in 1997 and which forms the basis for the Utility Guidelines issued by the USPTO, was directed to the issue of adequate written description for generic claims for nucleic acids when only a single species was disclosed in the specification. The case also addressed the issue of adequate written description for a species that was not disclosed, but which was related to the disclosed species. In this case, the three claims at issue were directed to cDNA encoding mammalian insulin, vertebrate insulin, and human insulin, while the disclosure was limited to an amino acid sequence for human insulin and a cDNA sequence for rat insulin. There was no disclosure of any other mammalian insulin species at the cDNA level.
The CAFC
drew a distinction between obviousness and adequate written description. It
cited case law holding that the claimed invention may, in fact, have been
non-obvious but still was not adequately described.[9] Further the Court stated that a claim may be
enabled, but still may not be adequately described. Therefore, it is clear that
a claim can be both non-obvious and fully enabled and still not be adequately
described.
Based on a
line of cases the CAFC cited in the Lilly
opinion, the test for adequate written description is whether the specification
conveyed the claimed invention with reasonable clarity so that the person of
ordinary skill in the art would have known from the disclosure that the
applicant was in possession of the claimed invention. Evidence of such
possession can take the form of actual reduction to practice, clear depiction
in detailed drawings or structural chemical formulas, or through written
description describing sufficient relevant identifying characteristics. Any of
this can be shown by words, structures, figures, diagrams and formulas, but not
by merely describing or arguing those features which may make the claimed
invention non-obvious or those technical details which may even enable the
invention. The Court stated that an adequate description of a DNA such as a
cDNA, would require a precise definition such as by structure, formula,
chemical name or physical properties, but not by a mere wish or plan for
obtaining the chemical invention (viewing, as the Court did, the claimed
nucleotide sequences as being chemical compounds).
In this
regard, the Court relied on the case of Fiers
v. Revel,[10]
in which an inventor had shown a method that may in fact have been
effective to produce the DNA sequence, but was not sufficient to support a
claim to “ a DNA sequence encoding protein X” because it did not disclose the
sequence itself. The Court made it clear that a name such as “cDNA” is not
sufficient and that the applicant had provided no distinguishing information
regarding the identity of the DNA, no sequence information and no relevant
structural or physical characteristic.
Since the Lilly Court used this rationale where a
nucleotide sequence is concerned and stated that only the sequence per se is
sufficient to describe a claim to the sequence, one could argue that Lilly might be limited to DNA sequences.
On the other hand, it is clear that the Court applied principles from the law
of the chemical art by treating DNA as a chemical entity, making it critical
that the DNA be characterized by its sequence and nothing less, and meaning
that Lilly in all likelihood will not
be limited to claims to DNA sequences.
The Court
stated the following, “ A written description of an invention involving a
chemical genus, like a description of a chemical species, requires a precise
definition, such as by structure, formula or chemical name of the claimed
subject matter sufficient to distinguish it from other materials” (citing Fiers). Therefore, to obtain a claim to
any chemical compound, including a nucleotide sequence, that compound must also
be described in detail in the specification in order to meet the written description
requirement.
The Court
also said that with respect to the claims to vertebrate or mammalian insulin,
the claims did not distinguish the claimed genus from other genuses except by
function and there was no definition of any structural features commonly
possessed by members of the claimed genus that would adequately distinguish
them from other genuses. The Court said that a definition by function does not
suffice to define the genus because a function is only an indication of what
the gene does rather than what the
gene is.
Therefore
at the USPTO, following this case and its predecessor cases, the focus is on
the following standards in the Utility Guidelines:
(1) Description of structure of
a representative number of claimed compounds; and
(2) Description of chemical or
physical characterization of a representative number of claimed compounds or of
the function of a representative number of claimed compounds (other than
binding to the protein at issue).
3.02 Written Description of
Compounds
Reach-through
claims, directed either to compounds per se that modulate the novel receptor or
to compounds identified by screening procedures against the novel receptor,
generally are not limited to specific species or to specific classes of
compounds, much less to specific structures. The claimed compounds may be, and
generally cover, a large and diverse group of molecules. These typically end up
including many thousands of large or small organic (we will follow an industry
convention that considers small molecules to be organic molecules having a
molecular of 500 of less) and inorganic molecules, antisense nucleic acids,
fragments of nucleic acids, peptides, ribozymes, antibodies and the like,
without actually disclosing core structures or representative numbers of core
structures or functions for any of them. Also typically, the specification may
disclose no modulator other than the single ligand of the example, which may
often be the natural ligand. Alternatively the specification may disclose
additional modulators but not have provided a sufficient number of modulators
such that a core structure (for example a Markush group) can be defined for one
or more classes of modulator.
In this
situation the USPTO would properly take the position that the claims are not likely
to comply with the written description requirement, citing Lilly and In re Ruschig.
[11]
However, where the specification does describe the structure of a
representative number of compounds that would provide a core structure, a claim
could potentially be obtained directed to compounds of this class. The USPTO
would have to be convinced that possession of the core structure is reasonably
predictive of the ability to modulate the receptor. To accomplish this is in
actuality an exercise in drafting a specification that, with respect to the
ligands desired to be validly covered by the claim, is as detailed as the
description would be in the situation where the applicant is claiming the
compounds per se in a conventional case, for example an organic medicinal chemistry
genus and species case. In other words, if you want a genus of compounds (that
interacts with your receptor) to be considered to be correctly described, look
to how it was done in a chemistry case. A detailed discussion of how to meet
the description requirements in chemical patent practice is beyond the scope of
this paper, but one can readily grasp the scope of what is considered to be
adequate description of a given genus by a study of issued patents in the area.
If you think di-substituted imidazoles block your receptor to induce
vasoconstriction particularly well, you should study some issued patents in
that art. A claim well-drafted in this manner will be much less likely to reach
through and be rejected on that basis.
Would the
CAFC likely uphold the USPTO’s application of Lilly to those situations in which claims cover compounds per se
that are ligands to a novel receptor, or cover compounds identified by
screening methods against the novel receptor, but the compounds have not been
disclosed as to specific formula, structure, or function, but rather by generic
class, such as organic molecule, inorganic molecule, ribozyme, and so forth?
Since the decision in Lilly was based
on a rationale derived from the written description requirement as applied in
the chemical art, it appears likely that the Court would apply the same standards, which it applied in assessing
adequate written description for the terms “vertebrate DNA” or “mammalian DNA”.
That is, without a specific structure(s), the disclosure would not be found to
adequately describe such a generic claim. With respect to the claiming of
compounds identified by screening, it could be argued that such a claim would
also be unlikely to meet the requirements set forth in Lilly for a generic claim unless screening conferred or resulted in
some identifiable generic structural limitation. At the very least, the
applicant would have to show that screening produced a representative number of
compounds, the term “representative” being based on some type of common
physical or chemical characteristic that would convey to the person of ordinary
skill with reasonable clarity that the inventor actually was in possession of a
particular genus of compound.
US Patent
Number 6,083,705 is an example of a claim drafting strategy attempting to get
around this problem, in which the claims are essentially drawn to compounds
identified by screening via couching this limitation as a process limitation
and tacking it onto the end of a screening claim. Screening claim 3 is as
follows:
“3. A process for determining whether a chemical compound is a
human α1c adrenergic receptor agonist which comprises
contacting cells transfected with and expressing DNA encoding the α1c
adrenergic receptor with the compound under conditions permitting the
activation of the α1c adrenergic receptor, and detecting an
increase in α1c adrenergic receptor activity, so as to thereby
determine whether the compound is an α1c adrenergic receptor
agonist.”
The
following dependent claim 31 was then submitted in an attempt to protect the
compounds identified by the method in the preceding claim:
“31. A method of preparing a pharmaceutical composition which
comprises determining whether a compound is an α1c adrenergic
receptor agonist or antagonist using the method of any of claims 3 through 6
and admixing said compound with a pharmaceutically acceptable carrier.”
The
‘705 patent is primarily directed to a novel receptor. A method of screening
claim, a claim otherwise allowable by the USPTO, contains a reach-through
process limitation. The method is written as a method for preparing a pharmaceutical
composition, which adds a step following the screening steps wherein a compound
is identified as an agonist or antagonist and is then mixed with a
pharmaceutically carrier. Such a claim could block others from methods of
treatment because in a method of treatment, a pharmaceutical composition is
needed. The disclosure however, does not contain description of specific
compounds or a representative number of compounds, but rather discloses that
genus of compounds that is based on the
collective ability of the compounds of the genus to be identified by a
screening procedure for a compound that modulates or binds to the receptor.
Thus, the disclosure is limited to what that compound does and not what that compound is.
Accordingly, under Lilly, it can
readily be argued that this claim is invalid and reaches through because the
compound that is the subject of the claimed limitation is a compound defined
not by what it is, but by what it does.
The USPTO
would allow the screening method for identifying compounds. But, can claim 31 be sufficiently described? If
we dissect claim it would appear to comprise the following elements:
A compound identified by screening,
which is admixed with a pharmaceutical carrier.
For simplicity, this element of the claim could be broken in two pieces. One piece is “a compound identified by screening against target X.” The other piece is a method for making a pharmaceutical composition “by admixing a compound with a carrier.” Based on properly applied practice of the USPTO as to the compound, neither element of this claim is adequately described in the specification. Since they are two elements, each requiring their own degree of description, one cannot simply get an additive effect by putting the two elements together, which would therefore not seem to be an adequate drafting strategy to overcome the deficiency. Based on the USPTO Guidelines, therefore, this claim should fail for lack of adequate written description since the compounds were not adequately disclosed. And it appears that there is a strong likelihood that the CAFC would adopt the same position because this claim to a method of making a pharmaceutical composition finds its point of novelty only in the compound.
3.03 Written Description of
Methods
The USPTO applies the same rationale to claims to methods of treating a disease by administering a compound identified by the screening procedure. This type of claim can be written more generically so that it is not directed to methods of treating a disease. It can be written as a “method for modulating the target protein by administering a compound under conditions suitable for modulation.” Such a claim (to a method of modulating) could potentially also reach through to methods of treatment, since it is a dominating generic claim.
The
position of the USPTO is that Lilly
extends to such method claims as well. That is, unless compounds have been
adequately described, a method for using compounds in the generic sense, either
compounds per se or compounds identified by screening, would not be adequately
described.
The
rationale for this position can be illustrated by the following hypothetical
claim based on Lilly:
“A method for assessing the
level of insulin in a mammalian cell by hybridizing a mammalian insulin cDNA to
a nucleic acid in said cell and assessing the level of hybridization between
said mammalian insulin cDNA and said nucleic acid in said cell. “
Since the
CAFC takes the position that a genus of compounds, i.e. mammalian insulin, is
not adequately described by the disclosure of one insulin species, using the
same rationale, this claim should, if issued, be invalid for lacking adequate
written description as to the genus “mammalian insulin”. The method, of course,
is per se old.
U.S. Pat.
No. 6,048,850 is an example of an issued method claim that is illustrative.
A method for selectively
inhibiting PGHS-2 activity in a human host, comprising administering a
non-steroidal compound that selectively inhibits activity of the PGHS-2 gene
product in a human host in need of such treatment.
The
disclosure in this patent generally refers to many genuses of compounds
including homologs of the gene, analogs of the gene, antisense compounds,
ribozymes, antibodies, peptides and small organic and inorganic molecules.
However, no specific compounds are identified and the specification does not,
therefore, provide a description of the structure of a representative number of
claimed compounds or a description of the chemical or physical characterization
of a representative number of claimed compounds (for example IR or NMR
characterizing data) or the function of these claimed compounds, other than by
generically stating that they would modulate the involved protein. Based on the
current position of the USPTO, this claim would fail to meet the written
description requirement for these non-steroidal compounds, following the
rationale in Lilly.
4 35 U.S.C. §112, 1st
Paragraph, Enablement
The second key flaw that makes a claim impermissibly reach through arises when the specification has been drafted in such a way that a person of ordinary skill in the art cannot make and/or use a reasonable number of claimed embodiments of the invention, i.e. across its claimed scope, unless that person has to apply an undue amount of experimentation, which is a requirement of the statute.
In general,
the USPTO applies the factors set forth in In
re Wands. [12] The claim at issue there was directed to a
method for detecting Hepatitis B surface antigen by using ultra-high affinity
monoclonal antibodies. Representative is claim 7.
“7. Monoclonal high-affinity IgM antibodies immunoreactive with
HbsAg determinants wherein said antibodies are coupled to an insoluble solid
phase, and wherein the binding affinity constant of said antibodies for said
HbsAg determinants is at least 109M-1 “.
The CAFC
indicated that the determination of what constitutes undue experimentation in a
given case requires the application of a standard of reasonableness, having due
regard for the nature of the invention and the state of the art as of the time
of filing the application. It indicated that the test is not merely
quantitative, since a considerable amount of experimentation is permissible if
that is what is routine in the industry or if the specification provides a
reasonable amount of guidance with respect to the direction in which the
experimentation should proceed. It cited the factors from Ex parte Forman, [13]
including: (1) the quantity of experimentation necessary in that art, field or
industry, (2) the amount of direction or guidance presented by the
specification, (3) the presence or absence of working examples in the
specification, (4) the nature of the invention, (5) the state of the prior art,
(6) the relative skill of those of ordinary skill in the art, (7) the
predictability or unpredictability of the art, and (8) the breadth of the
claims.
Applying these factors, the Court in Wands found that a considerable quantity of experimentation was necessary, but no more than that which was reasonably to be expected in that particular art or in that particular industry. The specification described how to make a hybridoma and screen the hybridoma for antibodies of a particular defined affinity. The Court described the nature of the invention as the development of monoclonal antibodies, involving screening hybridomas to find an antibody with a desired characteristic. It described the level of skill in the art as high. It also described that art as reasonably predictable, based on the fact that such antibodies had been for some time obtained against other antigens and there was no reason to believe that raising antibodies against the hepatitis antigen would be an exceptional case. The examples showed that there was a reasonable rate of success in obtaining the claimed high affinity antibodies.
Somewhat at the opposite end of the spectrum of allowability, are screening claims where the receptor is an orphan and no known way exists of activating the receptor and accordingly, providing a screening assay that can be made and used without more teaching.
4.02 Enablement of Compounds
The USPTO
has taken the position that the scope of receptor-ligand claims directed to
compounds that are enabled by the specification depends on the compounds themselves
that are in the specification. Receptor-ligand claims can generally be obtained
for the specific compounds disclosed. The position taken by the USPTO in the
most extreme case, that is, where only the original single ligand is disclosed,
is that the claims are not enabled across their entire scope. The reasoning
goes that such claims encompass a genus of compounds defined only by their
function (interaction with the receptor), and the relationship between the
structural features of members of the genus and their function has not been
defined. Therefore, in the absence of such a relationship that it is either
disclosed or would have been recognized, based on information generally
available to those of skill in the art, the person of ordinary skill in the art
would not know how to make and thus how to use compounds that lack such
structural definition. The USPTO takes the position that this deficiency cannot
be overcome merely because the person of ordinary skill could assay a compound
of interest “since one would have no knowledge beforehand as to whether or not
any given compound (other than those that might be particularly disclosed in an
application) would fall within the scope of what is claimed.”[14]
Accordingly, the USPTO concludes that it would require an undue burden of
experimentation to screen undefined compounds randomly for the claimed
activity.
How might a
court approach this question? A specific question is whether the term “make”,
with respect to compounds, necessarily requires something more than synthesis
(either in vitro or in vivo). It could be argued, for purposes
of enablement, that given the large number of compounds available,
“identification” of a compound meets the “how to make” requirement of section
112, first paragraph. “Make” could be
equated conceptually with “obtain” and, therefore, could be fulfilled by the
availability of such a large number of potential screenable compounds.
If a court
were to adopt such an interpretation, the question would then become how the Forman factors apply. That is, what type
of experimentation would be required to screen random compounds or classes of
compounds to find a hit. It could be argued that the drug screening art is
sufficiently developed to enable a reasonable rate of return for compound
identification. There is a diversity of compounds that exists in the art and,
if a court found the combinatorial library technology of sufficient maturity to
generate a diversity of compounds, the court might find such claims to be
enabled by virtue of the very existence of the sheer total number of compounds.
In the case of compounds identified by screening, what is the success rate
likely to be for obtaining compounds that would serve as antagonists, agonists,
or otherwise modulate a target?
The success
rate in screening not only combinatorial libraries but also a diversity of
compounds available to the person of ordinary skill in the art anecdotally
appear to be on the order of from between 0.1-1.0% from screening approximately
1 x 105 small molecule compounds, with success rates appearing to be
smaller for peptides. Accordingly, based on the maturity of the art of compound
screening and the level of return, it is possible based on the fact scenario in
Wands that such claims could be
enabled. Applying the Wands factor to
this case, we can conclude that (1) although a high quantity of experimentation
is required, this is expected in the industry; (2) screening of compounds that
modulate targets has been reasonably successful in the industry; (3) the level
of skill in the art is high; (4) the target is not one that would be subject to
any particular drawbacks that would distinguish it from any of the other
targets for which modulators have been found by screening; (5) the
specification directs the person of ordinary skill in the art to screening
processes if such were not already routine in the art. Certainly, working
examples would help. Extrinsic information could be presented to show that for
a given class of compound, a reasonable success rate could be predicted.
However, this would only be a reasonable success rate for the category of compounds that the pharmaceutical industry characterizes as “hits” Such compounds are only in the rarest circumstances able to be progressed through the rest of the new drug product development pipeline to become a marketed drug someday. Instead the “hit” compound (or, as is more typically the case, several hits produced by a high throughput screen exercise) is studied by organic medicinal chemists and pharmacologists as to how the compound can be made more “drug-like”. Being “drug-like” involves factors such as aqueous solubility, ability to cross the lining of the gastrointestinal tract, potential for inducing a given metabolic enzyme level, and so forth. An otherwise promising hit may be as soluble as a brick. So modification chemistry is undertaken to produce homologs, analogs and close congeners. This involves extensive chemical modification work around the structural formula of the “hit” until a molecule with more desireable properties results, which is categorized as a “lead”. The “lead” is progressed and may or may not get past efficacy and toxicity indicators. The “lead” is what is most sought after in terms of one or more allowed compound-per-se patent claims. The point here is that a lot of organic medicinal synthesis has to take place, with only a limited amount of predictability about the outcome of optimizing the hit to produce a lead. Should the imidazole hit be alkylated at the 4 position? Should it be acylated? Should the alkyl substituent be aminated and the amine in turn be mono- or di-substituted? The permutations quickly become very large numbers. Is there any teaching on any of this in the specification? The less teaching there is, the less likely it is that the claim to a receptor and a genus of ligands is enabled across its breadth and it therefore reaches through.
In general,
the position of the USPTO is that this type of claim (i.e. “A monoclonal
antibody which recognizes the receptor having the sequence which consists
of SEQ. ID No:X…”) is patentable even if
specific antibodies are not disclosed. This position may have the blessing of
the CAFC under the case of Enzo Biochem,
Inc. v. Gen-Probe Inc.[15]
This case was directed to the issue of written description. The patent claimed
nucleic acid probes that selectively hybridize to genetic material of bacteria
that cause gonorrhea. The CAFC, which affirmed a decision from the Board of
Appeals, discussed why written description for the claimed invention was
inadequate. The CAFC found that the patentee had provided only vague details
about the nucleotide sequences, and did not identify chemical structures of
probes. However, of benefit to future application drafters, the CAFC cited the
Written Description Guidelines as stating that an applicant could show that an
invention was complete by disclosing functional characteristics coupled with a
known or disclosed correlation between function and structure. The CAFC also
directed readers of the opinion to the Written Description Guidelines in order
to learn how to comply with the written description requirement for a claim
directed to an isolated antibody capable of binding to antigen X “considering,
‘the well defined structural characteristics for the five classes of antibody,
the functional characteristics of antibody binding, and the fact that the
antibody technology is well developed and mature’”.
Note that the
name “antibody” can be said to describe a chemical entity. The claim simply
calls for the antibody binding to the antigen, This could be construed as a
chemical entity defined by its function and not its structure. The USPTO
appears to have made an exception for antibodies because the art of raising antibodies
is now so well-developed. In point of fact, the structure of an antibody to
which the Court refers is a generic one and the ability to bind is a
characteristic that can not be predicted in advance for any given antibody for
a given antigen. The drug screening process could be analogized to antibody
production. The antibody is not made directly from the target protein. It is a
preexisting molecule circulating in the body. The target protein is needed to
select the cell that produces the antibody. The produced antibody is then
identified by means of the target protein. Moreover, a myriad of different
antibodies can be raised against a specific antigen. It would be impossible to
even disclose all of them, or to provide a ”core” structure for specificity. In
the same way, the target protein is needed to bind or otherwise biologically
interact with a compound in a biological assay. That is how the compound is
selected among the many possible compounds. The question may come down to one
of reasonable predictability, ie., the USPTO does not consider the screening
process to reach the level of predictability that exists in the antibody art.
4.04 Enablement in Pioneering
Inventions
In closing
the discussion on enablement it is worthwhile noting that the CAFC has recently
held in the case of Plant Genetic Sys.,
N.V. v. DeKalb Genetics Corp.,[16]
that no special consideration is given to an invention that can be
characterized as a pioneering invention in this regard, and such an invention
will not be entitled to a lower standard of enablement under 35 U.S.C. § 112,
first paragraph. Some practitioners may recall previous cases of the CAFC and
its predecessor court, the Court of Customs and Patent Appeals, that suggest
that lower standards of the rules of patentability should be accorded to
pioneering inventions, but the Plant
Genetics opinion characterizes those passages as dicta and not case law.
5 Obviousness
A brief
word is also in order about how obviousness can create reach-through claims. A
broad claim drawn to any and all ligands that bind to a newly claimed receptor
will be very hard to obtain in the context of the prior art. This is so since
in most instances there will already in the prior art have been at least one
drug to treat the disease and which inherently acted through the receptor, even
though the mechanism had not been fully understood. An old principle of pharmaceutical patent law
says that discovery of the underlying mechanism of action of an old drug does
not entitle the discoverer to a new patent that would dominate the
administration of the old drug. This does leave open the possibility though of
the discovery of a receptor in the pathophysiology of a disease for which there
was no prior drug in use that acted at that receptor. This situation may, all
other things being equal, get around the prior art problem, particularly if
submitted in the form of a method of screening or an assay claim.
Can a patentee ever recover damages for infringement that would include the award of a royalty on the sales of a drug discovered or validated through the unauthorized use of a patented target? This question has occupied the thoughts of inventors, investors, executives and attorneys in the biotech and pharmaceutical sectors for many years now. Thousands of patent applications were filed in a hope that a theory would emerge that would grant such damages, perhaps even trebled for willful infringement. This section examines the rules of US damages law as they would apply to a patentee seeking an award of reach-through damages from a US District Court.
A court’s authority to award damages arises solely from the Patent Act. 35 U.S.C. § 284 states “Upon finding for the claimant the court shall award the claimant damages adequate to compensate for the infringement, but in no event less than a reasonable royalty for the use made of the invention by the infringer.” The award is not discretionary, it is mandatory if infringement was found. There is no such thing as de minimus infringement or an infringement so small that no award would be made on that basis (although as a practical matter no rational patentee would engage in the expense of patent litigation where it was known beforehand that the infringement was relatively small). But the damages have to be compensatory, not punitive. The patentee must show that it was actually damaged, and how it was damaged. The court has to ask “had the infringer not infringed, what would the patentee have made?” Aro Manufactoring Co. Inc. v. Convertblel Top Replacement Co.