July 1, 2009 (Vol. 29, No. 13)
Y. Philip Zhang
Recent Court Rulings Have Led to Confusing Conclusions that Demand Deft Oversight
Patent protection for pharmacogenomics innovations is facing significant challenges. U.S. patent law has undergone major changes recently due to a series of court rulings that will have significant impact on biotechnology innovations. Traps for the unwary are plentiful, and missteps can lead to failure to obtain the desired protection. This article explores some of the challenges to and practical approaches in pharmacogenomics patenting.
To be patentable, an invention must not be obvious to a person of ordinary skill in the art (a skilled artisan). The standard for determining obviousness underwent a momentous shift as a result of the Supreme Court decision in KSR v. Teleflex. In KSR, the Supreme Court found that the prior test for determining obviousness was too restrictive on the scope and use of prior art. A broader range of prior art is now available to the patent examiner.
For example, prior art that is directed at a different problem than what is solved by a patent applicant may be applied as long as it is within the area of prior art that a skilled artisan should have known and considered. The examiner may also apply common sense possessed by the skilled artisan and consider whether an invention is obvious to try.
In an April 2009 decision, In re Kubin, the Federal Circuit applied the KSR decision in the context of biotechnology patenting and overruled a long-standing obviousness standard for gene patenting. Patent applicants now face significant difficulties in patenting DNA sequences that encode known proteins. Those that discover new proteins may want to withhold public disclosure until the encoding DNAs are discovered and filed on.
On the flip side, a significant number of patents on DNAs that had been issued prior to the decision could be vulnerable to invalidity challenges. As a result, IP restrictions on certain diagnostic applications could potentially be lessened. Stakeholders should consult with counsel to devise approaches to identify and address these emerging risks and opportunities.
The patent law requires that a patent contains a written description of the claimed invention. The central objective is to describe the invention clearly enough so that it allows a skilled artisan to recognize that the inventor invented what is claimed. The written-description requirement presents an intricate issue to pharmacogenomics patenting.
One frequent pitfall is where the description provides inadequate representative species to support broad genus claims. In the highly competitive field of pharmacogenomics, one often needs to decide whether to file early with less than ideal support or to file later when more is established but risk losing out to competitor’s patent filing or public disclosure. It is important for the patent specification to support the full scope of the claimed invention with adequate representative species and embodiments.
Another common misstep is having insufficient support for a method of use claim in terms of exemplary compositions that possess the claimed functionality. In Ariad v. Eli Lilly, the asserted patents involve gene regulation and the NF-kB transcription factor. A critical issue is whether the patent provides adequate written description on how reduction of NF-kB activity can be achieved. The patent disclosed no examples of molecules that possess the claimed activity. In the end, the Federal Circuit found the patent invalid.
Patent drafters should bear in mind that the key consideration with regard to the written-description requirement is to know what the desired protection is and to sufficiently support it. During patent drafting and prosecution, patent counsel should be abreast of the scope of the available written support when crafting claims and presenting arguments before the patent office. Sufficient examples of representative species should be included to support the desired genus claim. Back-up claims, typically dependent claims that are directed to preferred species or embodiments, should be included in the event that the broader genus claim fails to gain allowance or is later held invalid in patent reexamination or litigation.
An invention must fit under one of the four categories of process, machine, manufacture, or composition of matter to be patent eligible. Abstract ideas, laws of nature, or natural phenomena are not patentable. Recent case law can have a major impact on the patenting of some aspects of biotechnology innovations such as pharmacogenomics and related diagnostic applications.
In In re Bilski, an en banc Federal Circuit substantially modified its previous standard for patent-eligibility determination for process claims. The court held that a process is patent-eligible subject matter if: (1) it is tied to a particular machine or apparatus, or (2) it transforms a particular article into a different state or thing. The new standard could make it difficult to broadly patent certain biotech methods, particularly those that relate to biological processes and mechanisms of action. The decision could mean that, to the extent a method claim covers all potential ways to influence a biological process, the claim has become a claim on a natural phenomenon.
It is still too early to assess the full impact of Bilski. For one thing, the Supreme Court on June 1 granted certiorari and will review the Federal Circuit’s decision during the next term. If affirmed by the Supreme Court, Bilski would make it easier for courts to invalidate claims that are directed to collecting or processing human genetic information.
Patent drafters should consider including physical components or processes such that a method claim is not simply a data-gathering or -processing exercise. If certain assays are preferred, consider multiple claims of “performing a Y assay for X” with different assay types, for example, rather than simply claim “assaying for X.” Where innovative sample collection or preparation is pertinent, such features could help. Data presentation, especially if it allows more advantageous use of the data, should be considered as well. Naturally, one should be cognizant of the business value of a claim with too many limitations.
An invention must be useful to be patentable. The utility requirement, as explained by the Supreme Court in Brenner v. Manson, stems from the U.S. Constitution and the congressional intent that to receive a patent monopoly the invention must have a substantial utility and a specific current benefit.
An area where the utility requirement has had a significant impact is the patenting of genes and gene products such as biomarkers and their uses. In In re Fisher, the Federal Circuit held that expressed sequence tags are not patentable unless the functions of the corresponding gene are known or identified. It is increasingly clear that such uses as research intermediates or tools for monitoring gene expression (or other for-further-research types of utility) are not enough where the underlying gene lacks substantial and specific utility.
It is often the case that a patent application on a therapeutic or a diagnostic invention is filed before any clinical trial result is available. In fact, in many cases no human trials have even begun before patent filing. In general, one does not have to show a definitive correlation between a particular activity and an asserted therapeutic use of a compound, nor is evidence needed of actual success in treating humans. What is required is a reasonable correlation between the biological activity and the use asserted in the claims.
If appropriate, one way to satisfy the utility requirement is to include evidence of structural similarity between claimed compounds and a compound known to have a particular therapeutic, diagnostic, or pharmacological utility. Also worth noting is that data from in vitro assays or animal models, if reasonably correlated to the particular therapeutic, pharmacological, or diagnostic utility, may be sufficient to satisfy the utility requirement.
It should be noted that the nature and amount of evidence needed depends on the claim sought by the patent applicant. If one claims that a compound is useful in “treating a symptom” of an incurable disease it is likely to need less evidence or support than where one claims that the compound is useful in “curing” the disease.
Similarly, claims directed to the use of broad classes of agents, or treatment for multiple conditions, will generally require more written support than claims that are directed to a particular class or condition. Patent counsel should be apprised of the desired claim scope and the available evidence that could support the asserted utility.
Y. Philip Zhang, Ph.D., J.D. ([email protected]), is special counsel
at Cooley Godward Kronish. Web: www.cooley.com.