THE WRITTEN DESCRIPTION REQUIREMENT - Patent law

The first part of §112 ¶1 provides that the specification “shall contain a written description of the invention.” According to current federal law jurisprudence, this requirement is separate from the enablement requirement, meaning that an analysisaimed at determining whether the requirements of §112 ¶1 are met requires a separateinquiry for both written description and enablement. In turn, there are two aspects goal is commercial in nature cannot be kept secret. In the chemical arts, analytical tools for reverse engineering a chemical composition are very powerful. In contrast, a process for making a chemical product could more readily be kept a trade secret; reverse engineering a molecule or composition might not tell the process by which that material was prepared. In the pharmaceutical and consumer health arena,myriad regulations would prevent one from marketing a product without providing significant information regarding the contents. For this reason, intellectual property beyond patents can be of critical importance for protecting a product; trademarks are a critical component of intellectual property for many branded products.
to the written description requirement. The first aspect of the written description requirement is of significant practical importance, especially for the patent practitioner,while the second aspect relates to a more controversial interpretation of the written description requirement arising in a couple of important and recent CAFC decisions.

The first aspect of the written description requirement is often (but not always) encountered during the patent examination process when claims are amended or added during patent prosecution such that they are different from what was in the original patent application. In order for these amended or newly added claims to meet the written description requirement, the patent application as filed must provide written description support for any of the newly added subject matter. This means that the patent specification as originally filed must be able to provide a level of support sufficient to demonstrate the limitations and elements of each and every added or amended claim. An attempt to add new subject matter to a patent specification after it has been filed, whether such new subject matter is added into the claims during prosecution or into the specification by way of attempting to correct the original specification, is not allowed if that subject matter is not fully supported in the original patent specification. Any attempt to make such an addition during the patent prosecution process should draw a “new matter” rejection, and a request will be made that the newly added subject matter be removed.

Before proceeding further, it might be helpful to briefly review the patent prosecution process to better envision how new matter rejections typically occur.When an applicant files a patent application, that patent application is good for all the subject material that it contains—whether the subject matter is in the patent disclosure or in the original claims as filed. During the course of patent prosecution, it is very common for the patent practitioner to amend claims, remove claims, and add new claims to the patent application. The ability to make claim amendments (or even substitute entire claims) is important because such changes may be necessary to respond to objections or rejections that a patent examiner raises during the examination process. The rejection of a claim can be made for many different reasons, such as over questions of novelty, obviousness, or even the written description requirement itself, but whatever the reason, the appropriate amendment of a claim, or substitution of the claim with anew claim may circumvent the patent examiner’s objection or rejection of that claim.here are additional reasons an applicant may wish to amend or add claims during the patent prosecution process. Since the patent prosecution process usually unfolds over a period of several years, the applicants for a patent often learn more about their invention after the application is filed, meaning they may have learned more about the critical features of their invention, what they are and how to exploit them,and wish to make sure that those critical features become claimed features of the invention. The patent applicant is allowed to do this provided the claims, as amended or added, are supported by the patent application as originally filed. For this reason (among others), patent specifications often contain very detailed listings of reagents,excipients, ranges, structural embodiments, and the like. The required written support can also come from the description of the invention and all relevant embodiments, or be taught from the examples, drawings, formulae, etc. Such detailed disclosures can help supply the support necessary to amend and add claims at a later date.

In a straightforward example of a written description requirement violation involving new matter, let’s consider the example where a patent specification is filed that discloses specifically only the molecules A and B. During the course of the patent prosecution, the inventor contacts the agent prosecuting the application and informs her that he has prepared an additional example, compound C, and wishes to also claim it in the application that was already filed. He asks if it is too late to add the additional compound to the patent claims. What do you think? If you said,“No way!” You’re right! This amendment will not be allowed. The application as filed must provide complete support for the invention claimed. In this example, if the inventor wished to get literal patent coverage for the new compound then a new patent application containing a written and enabled description of compound would need to be filed. If a new patent application or a continuation-in-part application is filed containing compound C which the earlier-filed specification did not include,then compound C will be considered to have been filed, for purposes of establishing a filing date of priority for that compound, as of the date the new application or continuation is filed. This of course can affect the patentability of compound C since it is possible that relevant prior art might have appeared between the filing of the two patent applications (including possibly the publication of the applicant’s own first application). In the situation just discussed, the inventor did not include compound C any where in the first patent application, and thus an attempt to claim compound C in that patent application will most likely fail for lack of written description support.

There would have been a different outcome if the inventor had included compound C in the original patent application, even if the compound were not part of the original claim set that was presented when the patent application was filed. Where compound C is included in the patent disclosure (but not in the claims), the compound can be claimed later during the patent prosecution process because the later-claimed subject matter (compound C) has the necessary written description support in the specification. As a result, compound C would be afforded a priority date from the date the application was filed since the compound was part of the patent application as of that date, even though it was not specifically claimed until later. For this reason,the initial claim set in a patent application or even a subsequent claim set introduced during patent prosecution is not necessarily a complete picture of the potential of that patent application to issue additional claims. Yes, it is the issued claims that are enforceable against potential infringers, but it is the entire specification that serves as a wellspring from which those claims arise. If we recall the discussion back in

Markush structures and compound examples

FIGURE :Markush structures and compound examples disclosed in the 10/690,732 application.

Claims issuing from first application.

Figure : Claims issuing from first application.

During the course of the patent prosecution, the applicants were successful in getting the two claims shown in Figure above issued in U.S. 6,992,424 on July 15, 2004.Before to the issuance of the U.S. 6,992,424 patent, the applicants file a continuation application (application number 10/999,621). In the continuation application, two more claims to the remaining subject matter (Figure above) were issued in U.S. 7,532,923on September 29, 2008.

For the examples illustrated in Figures 9.2 and 9.3, both sets of claims were supported in the original application as filed, but the claims that issued in each patent described a different genus of compounds, and moreover, they issued in different patents on dates separated by more than 4 years. Despite this fact, both sets of claims were supported by the original disclosure, and consequently, the priority date for

Claims issuing from the continuation application.

FIGURE :Claims issuing from the continuation application.

both sets of claims traces back to the original patent application: June 1, 2001.This example is meant to demonstrate the importance of the disclosure as well as the claims for determining what can be properly patented. As long as the disclosure meets the patentability requirements (including the written description requirement) as of the date that the disclosure is filed, claims to that subject matter can be made in a patent application that properly claims priority from the one or more provisional patent applications, regular patent application(s), continuation application(s), or divisional application(s) containing the requisite disclosure.
One aspect in all this is that the claims that are originally filed with the specification are part of the specification, and thus should not qualify for new matter rejections. This means that an applicant may submit a list of claims with the original specification and during the prosecution may withdraw one or more of the claims,only to resubmit them later in the prosecution of that same patent application (or one

Claim 1 from third patent application (second continuation).

FIGURE :Claim 1 from third patent application (second continuation).

of its progeny). The newly introduced claim should not draw a new matter rejection because it was in the original patent application; the patentee had literal possession of those claims at the time the patent application was filed. Those claims must still be enabled and the best mode taught, but there should not be any issues related to new matter.
However, let’s further assume from the previous scenario that our applicant wishes to add a claim to the Markush structure shown in Figure 9.4 in yet a third patent application (second continuation) that contains the identical disclosure as the previous applications.
Referring back to the patent disclosure in Figure 9.1, we see that there was no written description support for the Markush of formula C; it was not disclosed in the patent application as filed. Attempting to add the claim to the compound C in a continuation patent application and claiming priority to the original patent application filing date will fail. If the applicant wished to add the claimed subject matter of the Markush structure shown in Figure above, she would have to file a new patent application(or a continuation-in-part) for which her priority date would be the new filing date,and any analysis of patentability in view of the prior art would have to take the new filing date into account.
The example just shown for genus C was a fairly straightforward matter. However,the question of whether claimed subject matter complies with the written description requirement is often clouded by the fact that the claim language does not have to be literally mirrored in the patent specification. For example, some elements may be inherent and although they were not literally present in the specification, they can still be explicitly claimed, provided such an element or elements was necessarily present in the thing being claimed, even though its presence was not overtly recognized. A patent specification might be filed describing a step in which a material is combusted but the requirement that oxygen be required for the combustion is not stated explicitly in the original patent application. Nevertheless, it should be allowable to amend aclaim to the process so that oxygen is included explicitly as a required element in the claimed process since oxygen is an inherently present reactant (by definition) in the combustion process.
Corrections to the specification of obvious mistakes can be made when one of ordinary skill in the art would be able to recognize the mistake as well as the correction that is to be made. What does this have to do with new matter or a failure to support the change? If an applicant corrects an application that has already been filed, the result is not unlike amending a claim during the patent prosecution process.

The difference in the language in the claim from what is disclosed in the original specification must be fully supported by the specification as filed, or the application is new subject matter and needs to be part of a new patent application. A change to the specification is no different. However, where the mistake and correction are both obvious, then no substantive change has been attempted; typographical errors easily fall within this category. Similarly, where an U.S. application is originally filed in a language other than English and a subsequent English translation is filed that contains an error, that error can be corrected using the originally filed foreign language as support for the correction.

In some instances, a patent applicant may wish to include large amounts of information in the patent specification but to do so would make the specification impractically long. For example, it might be known from one or more prior art references that an extensive group of compounds can act as polymer formation initiators. Instead of explicitly listing each and every reagent in the patent specification, the applicant may choose to incorporate those reagents by reference to one or more prior art document(s)that contain the information. The act of incorporating another reference or portion of another reference, if done properly, will have the same effect as if the reference or portion of the reference were explicitly present in the patent specification.

These are the relatively easy examples that lie on either extreme: Either the added material was clearly new matter or not. As we saw previously, an attempt to add anew compound or Markush structure in a claim that was not disclosed in the patent application disclosure as filed will likely draw a new matter rejection. In contrast,obvious mistakes with obvious corrections can be corrected without introducing new matter into the claims or specification. In reality, chemical and pharmaceutical cases are seldom as simple as these two extremes. In chemical patent cases, the core substance is often a Markush generic structure that the applicant wishes to claim some sub combination or a chemical composition or process where the reagents or reaction variables (temperature, time, concentrations) are described in a patent specification that teaches acceptable ranges but where the applicant wishes to claim one or more sub ranges or overlapping range. As a first approximation, one should expect that any patent claims that lack literal support in the specification are likely to face an uphill struggle should the applicant wish to claim such material. Likewise,when one argues that literal support in a patent application can be found but that support requires cobbling together different pieces of the patent, there will also be major challenges.

These issues will be demonstrated by our next case, Purdue Pharma v. Faulding Services. The Purdue Pharma Company (Purdue) owns U.S. 5,672,360 (the ’360patent) which is drawn to methods of treating pain by the administration of opiates(e.g., morphine) on a once-a-day schedule. A competitor company, Faulding began marketing a long-acting morphine formulation that included instructions that it could be administered once or twice a day. Purdue brought suit against Faulding alleging that Faulding’s marketing of their product infringed the ’360 patent. Claim 1 (which was not present in the original patent application as filed but rather was included during patent prosecution) of the ’360 patent is presented here:

We claim:
1. A method of effectively treating pain in humans, comprising orally administering to a human patient on a once-a-day basis an oral sustained release dosage form containing an opioid analgesic or salt thereof which upon administration providesa time to maximum plasma concentration (Tmax) of said opioid in about 2 to about10 hours and a maximum plasma concentration (Cmax) which is more than twice the plasma level of said opioid at about 24 hours after administration of the dosage form, and which dosage form provides effective treatment of pain for about 24 hours or more after administration to the patient (emphasis added).

In the District Court where the infringement action was heard, it was held that Faulding’s production and sale of their drug did infringe the asserted claims of the ’360patent but that those claims were invalid because they lacked the written description requirement required by §112 ¶1. On appeal, the CAFC panel hearing the case clarified the issue as whether the claim limitation “a maximum plasma concentration(Cmax) which is more than twice the plasma level of said opioid at about 24 hours after administration of the dosage form” (Cmax/C24 _2) was adequately described in the patent specification as filed. Noting that the District Court did not find support for that limitation in the patent specification as filed, the CAFC cited the standard level of support required as being that one of skill in the art, upon reading the specification,“must immediately discern the [claim] limitation at issue.”

On its behalf, Purdue argued that support for the critical limitation Cmax/C24 _ 2was found in the specification. In particular, Purdue argued that the portions of the specification reproduced below, describing their discovery as providing “24 hour oralopioid formulations which do not exhibit a substantially flat serum concentration curve” supported their specific claim language requiring a Cmax/C24 _ 2. Excerpt from specification:

They buttressed their contention that one of skill in the art would understand that “do(es) not exhibit a substantially flat concentration curve” means a Cmax/C24 _2 by providing evidence in the form of examples from the scientific and/or patent literature to make their case. The essential thrust of their arguments was that one of ordinary skill in the art would find their specification language equivalent to what they were trying to claim. Unfortunately for Purdue, none of the examples provided were persuasive to the CAFC because none of those cited references had the type of specificity needed to prove the correlation between the specification language and the claim language. For example, one of the cited references taught that a morphine formulation with a peak-to-trough morphine ratio of 60% to 100% was “substantially flat.” However, the CAFC pointed out that the reference was silent on the question of whether a peak-to-trough ratio of more than 100% would be not substantially flat; the CAFC was unwilling to accept Purdue’s implication from that statement. Purduefurther argued that the specific examples contained in the specification supported the more generic relationship as claimed. In this regard, the CAFC noted that while two examples out of the seven presented in the specification possessed the claimed trait, other examples in the specification did not. More important, there was nothing in the specification that would indicate that this relationship was a feature of their invention;there simply was no highlighting of the relationship at all, and nothing to indicate to one of ordinary skill that this specific relationship was their invention.

Citing a much earlier federal court case, the CAFC made the analogy that finding support for claim language in the specification was something akin to finding one’sway through a forest. Clear blaze marks needed to be left on the trees to point the way. The CAFC explained that it does not suffice to provide a forest of trees in the specification and then later pick out a single tree and say “here is my invention.”The direction for picking out those single trees must be present in the original specification. Any attempt to claim particular trees or groups of trees retrospectively that are not marked prospectively is overreaching by the patent applicant. Even where specific examples are present in the invention that do possess certain traits or numeric relationships, this does not, by itself, allow one to later claim any of those ratios,unless the specification clearly identifies the later claimed ratio as a clear feature of the invention. Regarding Purdue’s attempts to claim the key relationship from the mere presence of two examples that inherently contain the relationship, the CAFC summarized Purdue’s attempt as picking “a characteristic possessed by two of their formulations, a characteristic that is not discussed even in passing in the disclosure,and then making it the basis of claims that cover not just those two formulations, but any formulation that has that characteristic. This is exactly the type of overreaching the written description requirement was designed to guard against.”

The Purdue case can be described as a failed attempt to claim a broader generic relationship from specific examples described in the specification. The desire to claim subject matter broader then what is supported in the specification can also occur in cases that are more chemical in substance. In particular, one can envision different scenarios where an applicant describes specific examples of a process, amolecule, or a composition. In the case of a process where an actual example is provided, several variables (reaction time, temperature, solvent volume, etc.) will beset to specific values, the values that the process was actually conducted at. However,absent additional language in the original specification (including the originally filed
claims) explaining how such variables might be modified, a later attempt to broaden the examples by including ranges for any of the variables will likely fail for lack of proper support. For example, consider a patent application as filed that describes a specific example of a chemical process in which a reaction is heated at 30◦C. Absent the description of a temperature range in the original specification, the applicants would not be allowed to add claims to the process covering a range, for example,of 20◦ to 60◦C. An attempt to add such a claimed range during the prosecution of the patent should be rejected as improperly claiming new matter. However, if the original specification disclosed the process examples properly spanning a range of
20◦ to 60◦C, either in the disclosure or in the originally filed claims, then a new matter rejection is not appropriate.

A similar situation occurs when an original specification contains single molecule(species) examples only, and the applicant nevertheless attempts to add claim(s) to a broader genus that would contain one or more of the presented examples. For example, the disclosure of the compounds in the original specification as filed shown in Figure below would not provide sufficient support for adding a claim during the patent prosecution to the new Markush genus (also shown in the figure).Not only might applicants want to broaden their original filing by adding claims that exceed the scope of the originally filed patent application, the reverse situation can also occur; an applicant might attempt to add or amend claims to subject matter in a way that results in a narrower claim than what is supported by the original specification. Perhaps somewhat counter intuitively, this scenario will also fail for inadequate written description support.We previously discussed an applicant’s attempt at inappropriately broadening a single disclosed temperature for a process step of30◦C into a previously undisclosed range of 20◦ to 60◦C. In the reverse scenario, the specification would describe a range of 20◦ to 60◦C, and the applicant would attempt

Species disclosed in application and later attempted Markush claim.

FIGURE :Species disclosed in application and later attempted Markush claim.

Disclosure of genus does not provide written description support to species.

FIGURE :Disclosure of genus does not provide written description support to species.

to include a claim limitation containing a single temperature point (30◦C) that falls with in the range. The amendment would fail because the applicant is trying to add new matter to the patent application, even though the claim is narrower than the range disclosed in the specification.

Likewise, the disclosure of only the genus shown in Figure above in the original patent specification would not support the later attempt to separately claim any of the species that were not originally disclosed (the reverse of Figure 9.5). The broader disclosure of either a range or a Markush does not, by itself, allow one to separately claim the sub combinations or species components that make up the range or Markushstructure. The forest and trees analogy from the Purdue case is most appropriate in those circumstances since the original specification can be thought of as disclosing a generic forest and the later added claim as attempting to pick out a single tree.

This discussion begs the question of why an applicant might wish to narrow his description during patent prosecution to beyond what he originally filed. Where as the advantages of a broad claim are more intuitively obvious in terms of a broadened claim’s greater area of intellectual property real estate, the ability to narrow a claimcan serve several critical purposes as well. First, a claim as filed in the initial specification might not be valid in view of certain prior art, whereas a narrower claim might avoid the effect of that prior art. One scenario where this can occur is during the course of patent prosecution when the applicant becomes aware of prior art that renders his earlier filed claims not novel, whereas a narrower claim might avoid that problem .Another area where support for narrowing claim amendments is put into issue is in the arena of patent interferences. Sometimes when attempting to provoke an interference,one or more of the prospective parties to the interference will need to change his patent

Claims to Markush and species.

FIGURE :Claims to Markush and species.

claim language to match the language of the other patent application or patent so that an interference can be properly declared.
To tie these new matter issues together and place them into a patent prosecution context, let’s return to our previous Markush example from this section but augment the facts slightly. Envisage the scenario where the applicant has filed, in his original patent application, claims to the Markush and species shown in Figure above. No other Markush structure was included in the originally filed patent application. Very soon after the applicant files his original patent application, a competitor’s patentapplication publishes with the compounds shown in Figure below disclosed.

A quick look at the competitor patent application’s cover sheet indicates that the original application, from which the published patent application properly claims priority to and which contains support for the claimed compounds, was filed about

Intervening prior art species.

FIGURE :Intervening prior art species.

Amended Markush structure.

FIGURE :Amended Markush structure.

18 months before its publication. The applicant realizes that he probably can no longer claim the Markush genus because each one of the species from the competitor’s patent application falls within the scope of the Markush genus that our applicant is claiming.

When the applicant begins prosecution of the patent application, the first thing he does is submit an amended claim set where the claim to the Markush of formula I is canceled and the new Markush, shown in Figure above, is introduced as claim 3.
Why would the applicant wish to make the amendment? First, if he doesn’t, then the originally claimed Markush of claim 1 would be anticipated (rendered not novel)as it is currently written; recall that a single earlier disclosed species falling with in a later claimed Markush will anticipate the later claimed Markush structure. By redrawing the Markush definition carefully, it is possible to exclude the anticipating species while minimizing the amount of claim scope lost. Unfortunately the applicant in this case is almost certain to get a new matter rejection because at the time of filing, the applicant did not have possession of the later added Markush of claim3. Absent the necessary direction in the specification as filed, one of ordinary skill in the art would not have envisioned the Markush genus as added in claim 3. Moreover, it did not matter that the added claim was to a narrower genus of compounds; the written description requirement applies the same regardless of whether the subject matter is narrowed or expanded.
It is reasonable then to ask how the applicant might have avoided the problem of not being able to narrow his genus scope. While no plan is failsafe (especially in patent law), an applicant can increase his ability to claim varying, narrower slices of his Markush structure by providing support in his original specification as filed for those narrower embodiments. In the example we just analyzed, the applicant might have provided a number of alternate Markush structures of varying scope in the original patent application. Even without a crystal ball, the provision of narrowing Markush embodiments will increase the probability of finding fallback positions that are broader than the species embodiments but possibly still narrow enough to avoid intervening or non-discovered prior art.

For purposes of quick demonstration, let’s briefly return to the problem we just discussed in order demonstrate how the inclusion of various intermediate Markush structures might have helped the applicant. A reasonable way to narrow the Markush structure would be to provide a couple of embodiments wherein the species examples are still covered by each one of the included genera but where additional Markush scope is progressively whittled down. For example, the set of Markush embodiments might have been provided in the initial patent application as filed (either in the claims, the disclosure or both) as shown in Figure below. If the applicant in the previous example had provided the second and third embodiments shown in the figure, he would have been able to present alternate Markush structures that would not read on the prior art species disclosure while still providing broader coverage than the individual species embodiments. Note that these genera can be constructed rather systematically without having to see into the future but rather based on narrowing the broadest Markush so that alternate views of the invention will be provided in the specification as filed so that they can be claimed later if necessary.

Beyond the prior art issues like those just discussed, the presentation of the more narrowly tailored genera can also help avoid potential enablement issues. Although we will be discussing enablement in depth in the next section, suffice for now to appreciate that what is claimed must be enabled throughout the scope of the claim.It is not surprising that it is often more difficult to demonstrate sufficient enablementfor a broader claim than for a narrower claim because there is more subject matter to enable. By providing additional claims of varying scope, a failure to enable the

Progressively narrowed embodiments.

FIGURE :Progressively narrowed embodiments.
broadest genus does not necessarily lead to a failure to enable alternate descriptions of varying scope.
Although the written description requirement is most typically discussed in the context of when an applicant attempts to amend claims or the patent specification after the patent application has already been filed, there is a second, broader aspect to consider. This second aspect of the written description requirement relates not to the amendment process per se but rather to whether the claims, even where present in the original specification (such that there is no new matter issue), have written description support in the patent application. This written description requirement is separate from the new matter context discussed earlier, and is often discussed in the context of a separate requirement that the inventors demonstrate that they have “possession of the invention” at the time the patent application is filed by the disclosure in the specification itself. Asking an applicant to demonstrate possession of the invention is to require that she actually invented what she is claiming. An applicant could claima broad genus of compounds or compositions that are well within the ability of one of ordinary skill in the art to make and use, without the inventor having made any actual examples of the compounds that are being claimed. The synthetic techniques available to one of ordinary skill in the art are quite extensive, and viable schemes to many previously unprepared compounds can often be rapidly generated without resorting to extraordinary creativity or skill. If an applicant were to claim abroad genus of compounds or compositions as well as a method for identifying the utility of those compounds or compositions, the enablement requirement could arguably be met if acredible assertion of utility has been made. You might have an almost visceral reaction against such broad claims, even though they might arguably be enabled. One’s instincts should allow one to sense that broad claims need broad support (and vice versa) even if they are enabled. One should n’t be able to reap what she has not sowed.A recent case heard on appeal by a panel of the CAFC, Univ. of Rochester v.G.D. Searle & Co. provides an example of a written description, possession of the invention controversy. Although this is a pharmaceutical science rather than a purechemistry case, the general discussion in the opinion is instructive for its reflection of the current judicial policy values in regard to whether a broad claim finds adequate written description support in the specification (outside of the new matter context already discussed).
Traditional nonsteroidal anti-inflammatory drugs (NSAIDs) such as ibuprofen, naproxen, ketoprofen, and aspirin exert their effects by inhibiting cyclooxygenaseenzymes (cyclooxygenases) responsible for the production of prostaglandins. In the early 1990s scientists discovered that there were two distinct cyclooxygenases, which were subsequently labeled COX-1 and COX-2. It was determined that COX-1 was responsible for the production of prostaglandins in the gastrointestinal tract. The prostaglandins produced in the gastrointestinal tract by the COX-1 enzyme serveprimarily a protective role, and the inhibition ofCOX-1 is believed to result in stomachirritation and ulcers. In contrast, COX-2 has been shown to be responsible for the formation of prostaglandins involved in pain and inflammation. Selective inhibition of the COX-2 enzyme would thus reduce pain and inflammation but produce less gastrointestinal irritation than nonselective inhibition of COX-1 and COX-2.21
Scientists at the University of Rochester filed patent applications describing and claiming numerous aspects related to the COX-1/COX-2 pathway. One of the patent applications issued as U.S. 6,048,850 (the ’850 patent) and contained several broad claims to methods of administering to a person a non steroidal COX-2 selective compound. For the purpose of the present discussion, claim 1 is representative of the asserted claims from the ’850 patent and is shown here:
1. 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-2gene product to a human host in need of such treatment.(PGHS-2 is an acronym for prostaglandin H synthase, which is another name for theCOX-2 enzyme.)
The ’850 patent described the COX-1 and COX-2 enzymes as well as assays useful for distinguishing whether a compound was a selective inhibitor for the COX-2 enzyme.
The ’850 patent also described methods of using COX-2 selective compounds as well as their formulation, routes of administration, possible effective doses, and suitable dosage forms. On the day that the ’850 patent issued, the University of Rochester sued Pfizer, G. D. Searle, and related parties in a federal district courtalleging that Pfizer’s marketing of Celebrex and Bextra infringed several claims of the patent (including claim 1).

Pfizer alleged that the claimed invention was invalid for, among other things, failure to comply with the written description requirement. The District Court found forPfizer, holding that Rochester’s claimed invention was invalid for failure to comply with the written description because the Rochester patent neither disclosed a non steroidal compound that selectively inhibits the COX-2 enzyme nor provided any idea how such a compound could be made other than by “trial and error” research.

Instead of a specific description of the invention all the public received was a method of identifying and using compounds without any description of the compounds themselves.As the District Court explained in its decision,
Tellingly, . . . what plaintiff’s experts’ [sic] do not say is that one of skill in the art would, from reading the patent, understand what compound or compounds—which, as the patent makes clear, are necessary to practice the claimed method—would be suitable,nor would one know how to find such a compound except through trial and error. . . .Plaintiff’s experts opine that a person of ordinary skill in the art would understand from reading the ’850 patent what method is claimed, but it is clear from reading the patent that one critical aspect of the method—a compound that selectively inhibits PGHS-2activity—was hypothetical, for it is clear that the inventors had neither possession nor knowledge of such a compound.
The claims were therefore held to be invalid. The University of Rochester appealed the decision to the CAFC.

The CAFC upheld the District Court opinion, citing from an earlier decision:
“A description of what a material does, rather than what it is, usually does not suffice” to satisfy the written description requirement.24 The failure to disclose any actual compounds that would work for the claimed methods meant that the inventors failed to adequately describe their invention; they did not have possession of the subject matter that they were claiming. In effect, if the asserted claims had been held enforceable, then the patentees would have effectively captured any COX-2selective compounds, including compounds not yet made or even imagined at the time the application was filed. Claim 1 of the patent effectively covers the entire mechanism of action of what could theoretically amount to a very large and potentially incredibly diverse set of molecular structures, none of which was described in the patent itself. Instead of structural description of any compounds or substances that would accomplish the claimed methods, the public was offered the instructions as to how one could recognize such a substance if such a substance in fact existed. The essential test characteristic of the claimed method relied on a functional definition of any compounds to be used to accomplish the claimed method but did not provide the bridge between the required function and a structural corollary that any chemist could possibly understand. In the District Court decision, the federal judge ruling on the case considered the claimed method in this case as being similar to an attempt to claim a method of turning lead into gold:

In effect, the ’850 patent claims a method that cannot be practiced until one discovers a compound that was not in the possession of, or known to, the inventors themselves .Putting the claimed method into practice awaited someone actually discovering a necessary component of the invention. In some ways, this is reminiscent of the search for the so-called “philosopher’s stone,” eagerly sought after by medieval alchemists, which supposedly transmutes lead into gold. While the Court does not mean to suggest that the inventors’ significant work in this field is on a par with alchemy, the fact remains that without the compound called for in the patent, the inventors could no more be said to have possessed the complete invention claimed by the ’850 patent than the alchemists possessed a method of turning base metals into gold.

None of this is to say that a structural component must be explicitly laid out where one of ordinary skill in the art could readily equate the given function with a structure(which was not the case here). For example, if one listed a DNA sequence in a patent,it could be reasonable for one to also claim sequences that hybridize with the listed sequence since the base-pair complementarily is readily appreciated by one of skill in the art.

The broader issue in this case is not whether the patentees had made a discovery worth patenting but how far the federal courts were willing to allow the patentees go in exploiting that discovery. As a policy matter, we wish to reward inventors for their efforts and encourage innovation at the same time. If the patent-in-suit had been upheld, then its holders would have the right to tap into the royalty stream for anyCOX-2 selective drugs that might someday be discovered. Would the requirement of paying downstream royalties to the holders of a mechanism of action patent increase the pursuit of drugs acting on that mechanism? The answer is probably not. In the best case, companies would have the option of licensing the technology and would find it still financially viable to go after the targeted mechanism with novel therapeutic compounds. In the worst case, the owner of the mechanism of action patent would not allow others to work in the area at all, or alternatively, they might price its use so high that it was not financially viable for others to license. In these latter scenarios,the enforcement of the broad patent rights would decrease the actual amount of useful pharmaceutical research that could be done in the area.
Surely you are thinking that since the University of Rochester researchers had made such an important discovery that it merited some reward. As a matter of public policy, we undoubtedly wish to encourage these types of breakthroughs in basic research. As we have already come to appreciate, patent law is a delicate balancing act between encouraging inventors through the awarding of the exclusive right to practice the invention while at the same time ensuring that the net result is not harmful to the overall level of research and discovery. One could argue that “the patent law is the patent law” and that it should not matter what the policy might before any given circumstance, but the reality is that patent law is almost always subject to some degree of interpretation, meaning that there is often no strictly black or white answer to any particular issue before the courts. Where matters of interpretation must be made, the public policy behind the various interpretations cannot be ignored. In the instant case, we have a policy that says we should reward inventors, and we also have a policy that says we need to encourage innovation. Sometimes these go hand in hand, and at other times they are at odds. For example, when an inventor is rewarded with a patent that essentially can block or otherwise hinder all activity in a given area,the public policy might direct us to consider whether rewarding such inventors, as an inducement to future inventors, is worth the cost of blocking the application of that research in the present. Sometimes the answer is no; in such instances, the policies behind not allowing one patent to block the efforts of many might be honored by a court finding that the invention that was actually made is much narrower than the invention that was claimed. Accordingly, the interpretation of the written description requirement is sometimes the strategic ground where this policy meets patent law in the federal courts. This does not mean that the inventors do not deserve a patent but only that their patent should not overreach the scope of their discovery. Perhaps,for example, if claims directed to specific assays useful for determining whether a compound is selective for one of the cyclooxygenase enzymes had been asserted,their validity in the federal courts might have been upheld. In effect, it appears that is what their patent provided written description support for. In such a circumstance, one could plausibly argue that the courts would allow the University of Rochester to reap what it sowed without significantly hampering effort in the area of pharmaceutical research.


All rights reserved © 2018 Wisdom IT Services India Pvt. Ltd DMCA.com Protection Status

Patent law Topics