In the United States, a patentable invention must be four things: patent eligible, useful, new, and nonobvious. This chapter is devoted to these four requirements.
There are additional substantive requirements for the patent document itself that must be met for an invention to be patented. These requirements, particularly enablement, written description, and definiteness, present special challenges to those patenting biotech inventions and are the subject of the next chapter.
A patentable invention must be patent eligible. It must fall within one of four broad categories of invention listed in the patent statute—namely, composition, process, machine, and article of manufacture.
In the biotech and pharmaceutical fields, compositions include a broad swath of inventions. They can encompass, for example, small-molecule drugs, recombinant antibodies, drug formulations, recombinant cells and cell lines, genetically modified plants, and genetically modified animals. Processes can be methods for treating a disorder, diagnosing or prognosing a disorder, sequencing DNA, synthesizing or isolating a compound, and producing biologics using genetically modified organisms. Machines include anything from smartphone–based devices for imaging human organs or tissues to automated devices for performing high-throughput screening of compound libraries. Finally, articles of manufacture include, for example, diagnostic kits and medical implants, such as stents and artificial joints.
It is important to note here that patent eligibility and patentability are completely different concepts. Patent eligibility is a prerequisite—together with utility, novelty, and nonobviousness—for patentability. Thus, an invention that is patentable is entitled to a patent. An invention that is patent eligible, however, may or may not be entitled to a patent, since it may or may not also be useful, new, and nonobvious.
There is virtually no limit to the types of invention that can be patent eligible. There is, however, one essential caveat: A patent-eligible invention must result from human effort. Courts have long ruled that it cannot be a mere discovery, abstract principle, or law of nature.
EXAMPLE 3.1
Scientist A discovers Protein X, a surface marker on human melanoma cells. Starting with this discovery, Scientist A performs further research and files a U.S. patent application.
Claim 1 of the application provides a recombinant antibody to Protein X that binds to and lyses human melanoma cells.
Claim 2 provides a method for treating a human melanoma patient by administering to the patient the recombinant antibody of claim 1.
Finally, claim 3 provides Protein X.
Claims 1 and 2 are directed to a composition of matter and a method, respectively. Each results from human effort and would likely be considered patent eligible (independent of whether each would also be considered patentable).
Claim 3, on the other hand, is directed to Protein X per se. However scientifically important its discovery may be, Protein X, without more, is not patent eligible. It exists in nature and always has. Scientist A elucidated that fact but did nothing to bring Protein X into being.
The patent statute lists the four categories of patent-eligible inventions in scientifically neutral language. Over the past several decades, the courts, particularly the U.S. Supreme Court, have played a pivotal role in defining what these categories mean in the context of biotechnology.
In 1980, the Supreme Court issued its seminal patent-eligibility decision in Diamond v. Chakrabarty. This case involved a bacterium that was genetically engineered with exogenous DNA, permitting it to digest hydrocarbons present in crude oil. These genetically engineered bacteria were of value in helping to clean up oil spills. The issue here was whether genetically engineered organisms are patent eligible. The court held that they are. This decision helped open the door to countless recombinant organism patents, ranging from Chakrabarty’s engineered bacterium to Harvard University’s OncoMouse model for human cancer and Monsanto’s glyphosate-resistant Roundup Ready seeds and plants. The biotech industry fondly regards the Chakrabarty decision as a catalyst for its explosive growth in recent decades.
More recently, the definition of that which is patent eligible in the United States has changed in ways many consider unhelpful. Between 2012 and 2014, the U.S. Supreme Court, on three occasions, narrowed the definition of what is patent eligible.
Its 2012 decision in Mayo Collaborative Services v. Prometheus Laboratories concerned a patent for a method for performing a type of personalized medicine. The method involved administering a drug to a subject and then performing analyses, all with the goal of optimizing the drug’s therapeutic effect. The court held that a diagnostic or therapeutic method is not patent eligible if it does nothing more than generally instruct one to “apply the natural law” or if patenting it unjustifiably “forecloses more future invention.” Many in the therapeutic and diagnostic fields have severely criticized this decision as impeding health care innovation and development. Particularly troubling to innovators and attorneys alike is the ambiguity of the court’s language, the court’s apparent conflation of patent eligibility with the novelty and nonobviousness requirements for patentability, and the overall legal and commercial uncertainty this decision creates in the diagnostic and therapeutic industries. As with all Supreme Court decisions, it will take many years for the contours of the Prometheus decision to further take shape. Without congressional intervention or the Supreme Court’s overruling its own decision, it will likely be a long while before Prometheus’s full effect on the health care industry is understood.
The following year, in Association for Molecular Pathology v. Myriad Genetics, the court dealt with Myriad Genetics’ patents to isolated human breast cancer–associated genes BRCA1 and BRCA2 and their fragments. The question facing the court was whether isolating naturally occurring DNA, without doing more, renders the isolated DNA patent eligible. The court held that it does not.
In its decision in Alice Corp. v. CLS Bank, the court followed the reasoning of Prometheus and earlier decisions and raised the bar for software patent eligibility. The Alice decision related to computer-based methods for mitigating settlement risk. The court held that such methods are not patent eligible, since they are merely an abstract idea—namely, intermediated settlement. The fact that the methods are computer based was not enough, according to the court, to confer patent eligibility on “that abstract idea.” Although the patents in Alice Corp. were unrelated to biotechnology and pharmaceuticals, this decision affects software patents generally. An increasingly large number of biotech- and pharmaceutical-related inventions are software based, and this decision could affect these industries significantly.
Given this new judicial landscape, it is now far more difficult to establish the patent eligibility of isolated biomolecules, medical software–related inventions, and many types of diagnostic and therapeutic methods.
A patentable invention must have at least one use. That is, it must have a well-established utility. This happens when the invention satisfies two criteria.
First, based on the invention’s characteristics, the invention’s utility must be immediately apparent to one of ordinary skill in the relevant scientific field (also referred to as the art). Second, the utility must be specific, substantial, and credible. In the case of a biologic drug, for example, a specific utility could include a use for treating an identified disorder, such as diabetes or arthritis. A substantial utility could include a treatment having a therapeutically meaningful effect. And a credible utility could include a therapeutic effect that one of ordinary skill would find believable.
EXAMPLE 3.2
Scientist A, a chemist, synthesizes five organic compounds. These syntheses are performed to study organic reaction mechanisms rather than produce a new drug. None of these compounds is known to exist in nature. And none is reasonably expected to have any use, whether as a drug, a research tool, or something else.
Without more, each compound lacks utility. Among other things, each lacks specific utility, since no therapeutic, research, or other use is apparent. For at least this reason, none of the five compounds would be patentable.
EXAMPLE 3.3
Scientist A produces two transgenic mice. Mouse X, produced merely as a control, has cells containing a short, defined, noncoding exogenous sequence having no known function. Mouse X has no use as an animal model for testing human drugs or as any other research tool. Mouse Y, by contrast, has cells containing a human oncogene and is useful as an animal model for testing human antitumor drugs.
Mouse Y has a utility that is specific, in that it serves as an animal model for human antitumor drug testing. This utility is also substantial, since it is nontrivial. That is, Mouse Y has a “real-world” use. Finally, this utility is credible, since it is presumably believable to one of ordinary skill. Mouse Y therefore has a well-established utility. It would be patentable if it were also to satisfy the other patentability requirements.
Mouse X lacks the features of Mouse Y. For at least this reason, Mouse X does not have a well-established utility and thus would not be patentable.
Scientist A produces two radiolabeled fifty-residue single-stranded DNA probes, namely, Probe X and Probe Y. Each probe has a defined nucleotide sequence.
Probe X specifically hybridizes to a human BRCA1 gene having a defined mutation X correlative with breast cancer. Thus, Probe X has use as a tool for prognosing breast cancer.
Probe X has a utility that is specific, in that it serves as a tool for prognosing breast cancer. This utility is also substantial, since it is nontrivial. Finally, this utility is credible, since it is presumably believable to one of ordinary skill. Probe X therefore has a well-established utility and would be patentable if it were also to satisfy the other patentability requirements.
There is no known complementary region in the human genome, or any other genome, to which Probe Y specifically hybridizes. Simply put, no use for it can be articulated. For at least this reason, Probe Y lacks a well-established utility and would not be a patentable invention.
EXAMPLE 3.5
Scientist A discovers that Drug X inhibits glioma cell growth in vitro. In experiments on an accepted animal model for human glioma, Scientist A discovers that the administration of Drug X results in a reduced glioma growth rate.
In a subsequently filed U.S. patent application, Scientist A claims two methods, Method 1 and Method 2. Method 1 is for treating a human subject afflicted with glioma by administering to the subject an effective amount of Drug X. Method 2 is for eliminating glioma in a human subject afflicted with glioma by administering to the subject an effective amount of Drug X.
Method 1 has a well-established utility. It has a utility that is specific, since it treats a defined disease using a defined drug. This utility is nontrivial and thus substantial. Finally, this utility is credible, since treating glioma reasonably includes reducing its growth rate, and is presumably believable to one of ordinary skill. Method 1 therefore would be patentable if it also satisfies the other patentability requirements.
Method 2 lacks a well-established utility. It has a utility that is specific, since the cured disease and administered drug are clear. It has substantial utility, too, since curing a disease is never trivial. Without more, though, animal data showing a reduced glioma growth rate in response to Drug X do not support a reasonable belief that Drug X would eliminate the glioma. Thus, Method 2 would not be patentable.
Establishing credible utility can be difficult and time consuming for life science–related inventions, particularly therapeutics. Fortunately, at least in the United States, submitting evidence of operability to the Patent Office is permitted even years after a patent application has been filed in order to show an invention’s well-established utility.
EXAMPLE 3.6
Scientist A invents a new method for treating rheumatoid arthritis in humans using recombinant Antibody X (AbX). Three months later, Scientist A files a U.S. patent application claiming the method. Before filing, Scientist A does not have time to generate data using an accepted animal model for human rheumatoid arthritis.
When examining the application several years later (a topic covered in chapter 5), the examiner rejects the claims to the therapeutic method as lacking utility. Specifically, the examiner asserts that the claims lack credible utility since, in the absence of in vivo data, one of ordinary skill would not believe that the invention would work.
In response, Scientist A is free to submit evidence showing that the claimed method would likely work in humans. For example, the evidence could include data showing therapeutic success with AbX in an accepted animal model for human rheumatoid arthritis. Submission of such evidence would be permitted even if the evidence, or experiments used to generate it, were not described in the patent application.
A patentable invention must be new. That is, it must be novel. To be novel, an invention must not be known to the public in a way that violates the patent statute. For example, an invention is not novel if it has already been patented, described in a printed publication, publicly used, or on sale. As with all requirements for patentability, the novelty requirement is described in the patent statute in relation to the invention as claimed in a patent or patent application.
Public information that predates a claimed invention and can be cited by an examiner against a patent claim, such as a publication or third-party patent, is referred to as prior art. Prior art can destroy an invention’s novelty or render it obvious, depending on the facts.
In the patent statute, there are exceptions as to what constitutes prior art against a claimed invention. Perhaps the most important exception is for public disclosures by the inventor of a claimed invention. This exception provides a one-year grace period before such disclosure becomes prior art against the claimed invention. Grace periods are not universal. They are available in the United States but only in certain other countries. As an important aside, an inventor can disclose an invention via a confidentiality agreement (covered in chapter 18) without destroying the invention’s novelty.
EXAMPLE 3.7
Scientist A invents a new biologic drug, AbX, for treating head and neck tumors. Scientist A publishes a full description of AbX and its therapeutic use in a scientific journal. Within one year after the publication, Scientist A files a U.S. patent application claiming AbX and its therapeutic uses.
In the United States, Scientist A’s publication is not considered prior art against the claimed AbX and therapeutic method, since the publication falls within the statutory exemption to prior art.
It must be stressed that many other countries have no such grace period for an inventor’s own disclosure and instead use an absolute novelty system. For at least this reason, a U.S. patent application on an invention is normally filed before the invention is publicly disclosed. This way, an inventor can preserve foreign patent filing rights. The advantage of filing a patent application before disclosing an invention also applies when the inventor is foreign and the patent application is first filed in the inventor’s country.
If a prior-art reference destroys an invention’s novelty—or, more precisely, destroys the novelty of a claim to that invention—the reference is said to anticipate that claim.
To anticipate a claim, a prior-art reference must teach, or disclose, all elements of the claim. That is, the prior-art reference must teach all features of the invention as claimed, not merely some of those features.
EXAMPLE 3.8
Claim 1 of a U.S. patent application provides an isolated chimeric antibody that binds to gp120. Publication X, a prior-art reference, teaches a mixture of one thousand antibodies. One of those one thousand antibodies is a chimeric antibody that binds to gp120.
Publication X does not teach every element of the claimed invention. Namely, it does not teach a chimeric anti-gp120 antibody in isolated form. Publication X therefore does not anticipate claim 1.
An inventor’s own disclosure of an invention can destroy the novelty of a claim to that invention in the United States if the disclosure occurs outside the one-year grace period.
EXAMPLE 3.9
Assume the same facts as in example 3.7. That is, Scientist A invents AbX for treating head and neck tumors and publishes that invention in a scientific journal.
However, Scientist A does not file a U.S. patent application claiming AbX and its therapeutic uses until two years after the publication.
In the United States, Scientist A’s publication is considered prior art against claims to AbX and its therapeutic methods, since the publication falls outside the one-year grace period.
To anticipate a claimed invention, a prior-art reference must make an enabling disclosure of the invention. Chapter 4 covers enablement. For now, though, a prior-art reference makes an enabling disclosure of a claimed invention if, as of the patent application’s filing date, one skilled in the art could have made and used the invention without undue experimentation.
EXAMPLE 3.10
Claim 1 of a U.S. patent application provides a humanized antibody that specifically binds to pancreatic tumor cell Antigen X and lyses pancreatic tumor cells in vivo.
The specification provides ample description of precisely how to make such antibodies that not only bind to, but lyse, their target cells.
Publication X, a prior-art reference, recites the following language: “a humanized antibody that specifically binds to pancreatic tumor cell Antigen X and lyses pancreatic tumor cells in vivo.” However, unlike the patent application’s specification, Publication X does not provide any details about how to make or use the antibody. Instead, it merely names the invention.
Publication X does not anticipate claim 1, since it does not provide an enabling disclosure of the claimed invention. Put differently, based on Publication X, a skilled person could not have practiced the invention without undue experimentation.
A prior-art reference can anticipate a claimed invention by inherency. That is, if a prior-art reference teaches that which is in fact the claimed invention, the reference anticipates it. This is true even if the reference is silent on the invention’s features as they appear in the claims, and a skilled person would not have inferred those features from the reference.
EXAMPLE 3.11
Claim 1 of a U.S. patent application provides Compound Y.
Publication A, a prior-art reference, teaches Compound X and its use as an oral drug for treating a disorder in humans. It is a fact that the human body metabolizes Compound X to form Compound Y. It is also a fact that orally administering Compound X to a patient, per its intended use, necessarily results in the formation of Compound Y.
So, the disclosure of Publication A (i.e., Compound X’s administration to humans), when practiced, necessarily results in forming the claimed invention (i.e., Compound Y). Publication A therefore anticipates the invention of claim 1 by inherency.
A prior-art reference disclosing a known species of a broadly claimed invention anticipates the claimed invention. That is, when a claim provides a genus of compounds or other forms of invention (e.g., a genus of methods), a prior-art reference teaching even a single species of that genus destroys the novelty of that claimed genus.
EXAMPLE 3.12
Claim 1 of a U.S. patent application provides “a catalytic DNA molecule of at least forty-five residues in length, comprising (i) a central twenty-five-residue DNA-cleaving domain having consensus sequence 1, and (ii) first and second terminal domains, having a combined length of at least twenty residues, which are complementary to regions upstream and downstream, respectively, of the DNA sequence that the catalytic DNA molecule cleaves.”
Claim 1 encompasses a genus of catalytic DNA molecules. In other words, it encompasses a virtually infinite number of molecules, whereby each species of molecule has the required minimum length, the central consensus sequence, and the two terminal domains. The vast number of species results, of course, from the number of possible sequence and length permutations for the terminal regions.
Publication X, a prior-art reference, discloses a catalytic DNA molecule. This molecule (i) is fifty residues in length, (ii) has a central twenty-five-residue DNA-cleaving domain having consensus sequence 1, and (iii) has first and second terminal domains having a combined length of twenty-five residues and having sequences 2 and 3, respectively. Sequences 2 and 3 are complementary to regions upstream and downstream, respectively, of the DNA sequence that the molecule cleaves.
Publication X discloses a species of the claimed genus. It therefore anticipates the claimed invention, even though the disclosed species is only one of a vast number of species within the genus.
The converse is not generally true, and a prior-art reference teaching a genus does not typically anticipate a claim to a single species.
EXAMPLE 3.13
Claim 1 of a U.S. patent application provides a forty-five-residue catalytic DNA molecule having sequence X. This molecule includes (i) a central twenty-five-residue DNA-cleaving domain having consensus sequence 1, and (ii) first and second terminal domains complementary to regions upstream and downstream, respectively, of the target DNA sequence.
Claim 1 provides a species of catalytic DNA molecule, in that it encompasses a catalytic DNA molecule having one, and only one, length and sequence.
Publication X, a prior-art reference, discloses a catalytic DNA molecule of at least forty-five residues in length. The molecule has a central twenty-five-residue DNA-cleaving domain having consensus sequence 1. It also has first and second terminal domains having a combined length of at least twenty residues, which are complementary to regions upstream and downstream, respectively, of the target DNA sequence. No examples of the catalytic molecule are disclosed in Publication X.
Publication X discloses a genus encompassing the claimed species along with a vast number of other species. It does not anticipate the claimed invention.
An exception to this outcome occurs when the claimed species can be “at once envisaged” from a prior-art reference teaching a genus that includes the species.
EXAMPLE 3.14
Assume the same facts as in example 3.13. That is, claim 1 provides a forty-five-residue catalytic DNA molecule having sequence X and including a DNA-cleaving domain and first and second terminal domains. Again, claim 1 provides a species of catalytic DNA molecule having only one length and sequence.
Now, however, assume that Publication X makes a narrower disclosure. Specifically, Publication X discloses a forty-five-residue catalytic DNA molecule having (i) a central twenty-five-residue DNA-cleaving domain having consensus sequence 1, and (ii) first and second terminal domains having a combined length of twenty residues, wherein the first terminal domain has sequence 2 or 3, and the second terminal domain has sequence 4 or 5. The molecule of claim 1 corresponds to the disclosed molecule when the first and second terminal domains have sequences 2 and 4, respectively.
Here, Publication X discloses a genus encompassing a total of only four species. They are the claimed species (i.e., having sequences 1, 2, and 4) along with only three other species (i.e., having sequences 1, 2, and 5; sequences 1, 3, and 4; and sequences 1, 3, and 5). Publication X does not actually recite the claimed species of molecule. However, the claimed species is one of only four catalytic DNA molecules that can be envisaged from the disclosure of Publication X. It is therefore likely that the claimed species can be at once envisaged from Publication X and that this reference therefore anticipates the claimed invention.
To be patentable, it is not enough that an invention be patent eligible, useful, and new. It must also be nonobvious. Under the U.S. patent statute, an invention is not patentable if the differences between it and the prior art are such that the invention would have been obvious to a “person of ordinary skill in the art” as of the patent application’s filing date.
Determining whether an invention would have been obvious is a difficult and nuance-laden task indeed. Obviousness findings are always fact based. They depend on the invention as claimed, the time the invention was made, what was known then and by whom, and many other factors. For at least these reasons, obviousness determinations can be unpredictable, and an invention deemed nonobvious in one instance might be found obvious in another, if different facts are brought to bear.
In a sense, determining obviousness entails a journey back in time. It involves speculating as to what a hypothetical person having a defined level of skill in a defined scientific field would have thought about a hypothetical combination of known facts at a defined point in the past. In particular, it involves speculating as to whether these known facts would have led this person to make the invention with a reasonable expectation of success. Making this determination even more difficult is the prohibition against relying on the ever-so-human tendency to use hindsight—to impute knowledge of the present invention to one of ordinary skill in the past.
Judges and patent examiners alike must determine whether claimed inventions would have been obvious. This requires considering a number of defined factors in light of the available evidence.
The first consideration is the scope and content of the prior art. What exactly does the prior art teach, and how clearly does it teach it? The second consideration is the differences between the claimed invention and the prior art. Do the prior-art references suggest combining elements to arrive at the claimed invention? Third, what is the level of ordinary skill in the art? For example, what level of educational attainment (e.g., B.S. or Ph.D.), specialty (e.g., pharmacology, therapeutic antibody design, or genomics-related algorithm design), and level and type of work experience (e.g., five years specializing in small-molecule drug design) are required? Finally, secondary considerations, also known as objective indicia of obviousness, must be considered to the extent they apply to a given invention.
Perhaps the most important such consideration is unexpected results. A claimed invention would likely not have been obvious if, when invented, it would have surprised one of ordinary skill in view of the prior art. Put differently, a claimed invention would likely not have been obvious if, as of the patent application’s filing date and in view of the prior art, one of ordinary skill would not have reasonably expected the invention to succeed. A showing of unexpected results is often considered tantamount to determining nonobviousness.
Some other secondary considerations include the invention’s commercial success, a long-felt but unsolved need for the invention, and failure of others to arrive at the invention. Any of these factors, if proven, can help to establish nonobviousness.
Again, it is unpredictable whether a patent examiner or judge will find a claimed invention obvious. Each determination is a fact-based one. We can speak of future obviousness determinations only as being more or less likely to occur, and of certain facts as supporting or contradicting a finding of obviousness. Absolutes are rare.
With that in mind, the following examples give a flavor of the interplay between patentability’s nonobviousness requirement and the virtually infinite number of inventions possible in the biotech and pharmaceutical fields.
EXAMPLE 3.15
Claim 1 of a U.S. patent provides a method for treating rheumatoid arthritis in humans by administering monoclonal antibody X (MoAbX), a humanized antibody. MoAbX specifically binds to tumor necrosis factor alpha (TNFα) with high affinity.
Reference A, a prior-art publication, teaches a method for making a MoAb against a defined antigen. Reference B, a prior-art publication, teaches TNFα’s role in rheumatoid arthritis and the value of an anti-TNFα antibody-based approach to treating it.
Absent evidence to the contrary, a reasonable basis exists for concluding that the claimed stent would have been obvious over references A and B combined. That is, it is reasonable to conclude that one of ordinary skill, as of the patent application’s filing date, would have known to combine the teachings of references A and B to arrive at the claimed method.
Of course, this outcome could change dramatically depending, for example, on how a person of ordinary skill is defined, when the patent application was filed, or whether there is any other prior art teaching away from the claimed method.
EXAMPLE 3.16
Here, though, the claimed method is for treating humans refractory to other antibody-based rheumatoid arthritis treatments, and MoAbX specifically binds with high affinity to Epitope X on TNFα.
References A and B teach, respectively, a method for making a MoAb against a defined antigen and the value of anti-TNFα antibody-based approaches to treating rheumatoid arthritis. The discovery of Epitope X and its value as a therapeutic target is the inventor’s own. The prior art does not teach it.
Given these new facts, and absent evidence to the contrary, one could reasonably conclude that the claimed method would not have been obvious over references A and B combined. This is because one of ordinary skill would not likely have known to target Epitope X. Furthermore, even if Epitope X had been known, it is unlikely that one of ordinary skill, without more, would have reasonably expected MoAbX to succeed in treating patients with refractory rheumatoid arthritis. Presumably, no basis would have existed for reasonably expecting a refractory patient to respond to an antibody-based treatment targeting Epitope X.
Claim 1 of a U.S. patent provides a method for treating depression using Drug X at a specified dose.
Reference A, a prior-art publication, teaches Drug X and its use, at half the claimed dose, for treating bipolar disorder. Reference A also teaches that Drug X causes significant side effects. Reference B, a prior-art publication, teaches the failure of others to successfully treat depression using Drug X at doses half of the one claimed.
Absent evidence to the contrary, one could reasonably conclude that the claimed method would not have been obvious over references A and B combined. This is because the known failed attempts to treat depression using Drug X, as taught by reference B, would have dissuaded one of ordinary skill from attempting to treat depression using an even higher dose.
EXAMPLE 3.18
Claim 1 of a U.S. patent provides “a coronary stent comprising a stent framework coated with a composition of (i) a member of the Polymer X family, and (ii) an anti-atherosclerotic agent, such that the agent elutes from the stent over time.”
Reference A, a prior-art publication, teaches methods for making polymer-coated stents. Reference B, a prior-art publication, teaches drug-eluting stents made using members of the Polymer X family. Reference C, a prior-art publication, teaches therapeutic and prophylactic advantages of administering anti-atherosclerotic agents over extended periods of time.
Absent evidence to the contrary, a reasonable basis exists for concluding that the claimed stent would have been obvious over references A, B, and C combined. In this scenario, one of ordinary skill presumably would have been motivated to combine the teachings of references A, B, and C to arrive at the claimed stent. This is because the stent is made using a known method to incorporate a known category of drug having known advantages when administered over time (which elution from a stent permits). Here, the inventors did not overcome any technical obstacles, and the claimed stent does not behave in a way that is surprising over the prior art.
Here, though, the claimed stent is coated with a composition of (i) Polymer X1 (one of the one hundred members of the Polymer X family), and (ii) Drug Y (one of the one hundred types of anti-atherosclerotic agent), such that Drug Y elutes at a constant rate for at least ten years.
References A, B, and C teach, respectively, methods for making polymer-coated stents, drug-eluting stents made using members of the Polymer X family, and advantages of administering anti-atherosclerotic agents over the long term.
Given these facts, and absent evidence to the contrary, one could reasonably conclude that the claimed stent would not have been obvious over references A, B, and C combined. Presumably, this is because one of ordinary skill would not have reasonably expected the specific combination of Polymer X1 and Drug Y to yield this particular elution profile. Naturally, additional prior art teaching earlier failed attempts to achieve constant Drug Y elution from a stent, for example, would reinforce this nonobviousness argument.
EXAMPLE 3.20
Claim 1 of a U.S. patent provides a method for treating pain by administering to an afflicted subject a composition comprising Drug X and Drug Y.
Reference A, a prior-art publication, teaches Drug X and its use as an analgesic. Reference B, a prior-art publication, teaches Drug Y and its use as an analgesic.
Absent evidence to the contrary, a reasonable basis exists for concluding that the claimed method would have been obvious over references A and B combined. The claimed method combines two known analgesics to achieve a result that, in this scenario, does not overcome a pharmacological hurdle, such as an expected adverse reaction between Drugs X and Y.
Here, though, the dose of Drug X in the administered composition is 20 percent of the dose given when Drug X is administered independently, and likewise for Drug Y.
References A and B teach Drugs X and Y as analgesics, respectively.
Given these new facts, and absent evidence to the contrary, one could reasonably conclude that the claimed method would not have been obvious over references A and B combined. Presumably, the lower effective doses of Drugs X and Y are made possible because of the synergy between them. Let us assume that this synergy is a discovery underlying the claimed method and is not taught or suggested in the prior art. One of ordinary skill would not likely have expected a five-fold dose reduction of Drugs X and Y to succeed, even when these two drugs are administered in combination. In this scenario, there is no prior art reasonably supporting that expectation.
As with establishing utility, one can establish nonobviousness by submitting evidence of unexpected results and the like to the Patent Office after a patent application has been filed. This evidence typically includes prior art that teaches away from the claimed invention. Such prior art—for example, textbook chapters and scientific publications—discloses information (e.g., toxicity or experimental failure) that would have dissuaded one of ordinary skill from trying the claimed invention or reasonably expecting it to succeed. Evidence of nonobviousness also includes expert declarations and oral testimony establishing one or more factors supporting a finding of nonobviousness.
EXAMPLE 3.22
Scientist A invents a new peptide. Three months later, Scientist A files a U.S. patent application claiming the peptide. The claimed peptide has the structure [aa1-aa2-aa3]10, where aa1, aa2, and aa3 are defined amino acid residues. This peptide, when properly affixed to an antibody, extends the antibody’s in vivo half-life.
During examination of the application several years later (again, a topic covered in chapter 5), the examiner rejects the claims to the peptide as obvious. Specifically, the examiner asserts that the claimed peptide would have been obvious in view of prior-art references A and B combined. Reference A teaches [aa1-aa2-aa3]5 and its use for extending the in vivo half-life of a large protein when properly affixed to it. Reference B teaches the advantages of [aa1-aa2-aa3]-containing peptides as in vivo half-life-extending agents. In response, Scientist A is free to submit evidence of the claimed peptide’s nonobviousness. For example, the evidence could include a prior-art publication showing that other peptides having lengths greater than that of [aa1-aa2-aa3]8, when properly affixed to an antibody, create stearic hindrance and impede the antibody’s function. Depending on the facts, this publication could help to overcome the examiner’s rejection, since it presumably teaches away from the claimed peptide.