America-Israel Patent Law

A Nice Way To Start The Day

It's always nice to open up PAIR and see this:

Would be nicer if the appeal hadn't been necessary in the first place. And if the examiner got dinged (or had to reimburse the applicant) for this - it's not the first time this applicant has had to appeal with this examiner.

Still, the good guys won, which is not a given, so we're thankful for that.

Inherently Incomprehensible: Does the CAFC’s View of Inherency in In Re Kao Spell the End of Second Medical Use Patents in the USA?

The Federal Circuit’s May decision in In re Kao caught the attention of the patent prosecution community in general, and those of us working in pharma in particular, because it seemed to give some more contours to the obviousness doctrine post-KSR.  The decision involved three patent applications, all of which relate to aspects to Endo’s Opana ER^®, a commercially successful extended-release version of the opioid narcotic oxymorphone.  The Board of Patent Appeals and Interferences (BPAI) had rejected all three applications on grounds of obviousness; the CAFC reversed and remanded in one application (11/680,432) and sustained the Board’s rejections in the other two cases (12/167,859 and 11/766,740). 

It was the decision in the ‘432 application that grabbed the headlines.  Among other things, the CAFC said that it was wrong for the PTO to say that the dissolution profile recited in the claims of that application was obvious in view of the Maloney reference (WO 01/08661), since Maloney disclosed determining the dissolution profile using a test which was different from the test recited in the claims, and there was no evidence that the two tests were equivalent.  Interested readers can check the summaries in PatentDocs and PatentlyO for more information. 

What doesn’t seem to have attracted widespread attention, but did catch the eye of Endo’s attorneys (Jeff Lewis et al.), was the Board’s treatment of the ‘859 application and the CAFC’s upholding of that decision, in particular its application of the inherency doctrine.  Under the CAFC’s case law, this doctrine says that even if no one appreciated what the prior art taught, the fact that a product or process was inherent in the prior art means that the thing now being claimed is not new.   

The CAFC’s 2003 Schering v Geneva Pharmaceuticals decision provides a reasonable illustration of this doctrine.  In that case, the one and only claim of Schering’s patent application, a claim directed to a particular chemical chemical compound, was found to lack novelty: it was undisputed that the claimed compound, a metabolite, was formed in the human body whenever a person ingested a different, known compound, and that ingestion of that different, known compound by people had been taught in the prior art.  Thus, despite that fact that before Schering, no one had actually identified the metabolite, the claimed compound per se was inherently disclosed by the prior art (since the prior art taught how to make the compound – just administer the precursor compound to a person) and therefore the claim on the compound per se lacked novelty.  In that decision, the CAFC noted that Schering could have secured protection had its claims been worded differently, for example if it had claimed the compound in solid form or substantially pure form, neither of which were inherently disclosed by the prior art. 

Inherency is thus a basis for making a novelty rejection against a claim.  In contrast, a claim rejection on grounds of obviousness requires that, at the time that the invention was made, the claimed invention was obvious to one skilled in the art on the basis of the prior art.  Thus whereas a novelty rejection merely requires that the claimed thing existed, even if no one appreciated that existence – and therefore such a rejection can be based on grounds of inherent prior disclosure – an obviousness rejection requires that one of skill in the art appreciated what the prior art taught.  Without such appreciation, the claimed invention could not have been obvious at the time it was made.

Inherency came into play in that the claims at issue in the ‘859 application, claims 8 and 21, were found to be obvious – not anticipated – because a property recited in the claims was deemed to be an inherent property of oxymorphone.  Claim 8 reads,

8.         A method for treating pain in a human subject in need of acute or chronic pain relief, comprising the steps of:

(a) providing a solid oral dosage form comprising about 5 mg to about 80 mg oxymorphone or a pharmaceutically acceptable salt thereof in a controlled release delivery system with a release rate profile designed to provide an adequate blood plasma level over at least 12 hours to provide sustained pain relief over this same period, the system comprising a filer and a hydrophilic material, wherein oxymorphone is the sole active ingredient; and,

(b) administering the dosage form to the subject, wherein the oxymorphone C_max is at least about 50% higher when the dosage form is administered to the subject under fed versus fasted conditions.

It’s the last, emphasis-added clause which is critical. 

The only art relied upon by the BPAI and the CAFC in rejecting claim 8 was Maloney.  Both the USPTO and Endo agreed that Maloney does not actually teach that there is a different release profile for oxymorphone, or any other opioid, when the drug is administered to a fed patient versus when it is administered to a fasting patient.  And it’s uncontested that prior to the disclosure of this fact in Endo’s own patent application, one of skill in the art would not have recognized this different release profile.  Nevertheless, the CAFC found that this property was an inherent property of oxymorphone itself, and that since Maloney made it obvious to make and administer the composition recited in claim 8, the claimed method was obvious.  As explained by the CAFC,

The examiner rejected claims 8 and 21 as obvious in view of, among other references, Maloney. The examiner found that Maloney "teaches oral sustained release preparations of opioid analgesics" with the use of oxymorphone as a preferred opioid.

On appeal, the Board affirmed the examiner's rejection, relying exclusively on Maloney. The Board found that Maloney discloses a controlled release formulation with an opioid in amounts of 5- 100 mg and that oxymorphone is a preferred opioid. The Board found that Maloney further teaches using calcium sulfate (a cross linking agent), lactose (a filer), and hydrogenated vegetable oil (a hydrophobic material) in his formulation. Based on these disclosures, the Board determined that it would have been obvious to a person of ordinary skill in the art to formulate the claimed oral dosage form and to administer the form to the subject as claimed in the '859 Application.

* * *

Endo argues that Maloney does not expressly disclose the "food effect" limitation: "wherein the oxymorphone C_max is at least about 50% higher when the dosage form is administered to the subject under fed versus fasted conditions." Endo asserts that the Board erroneously relied on the teaching in the specification of the '859 Application that the claimed "food effect" was a property of oxymorphone and that Maloney inherently disclosed the limitation. Endo argues that an obviousness rejection can only be based on what is known by those of skill in the art at the time of the invention, and there is no evidence in the record that anyone recognized the claimed food effect at that time.

The Office responds that substantial evidence supports the Board's finding of inherency. The Office further responds that the Board's reliance upon the specification of the '859 Application to support this conclusion is entirely proper. Further, the Office responds that inherency is indeed a part of the obviousness inquiry.

This court agrees with the Office. Substantial evidence supports the Board's finding, based upon the specification, which confirms that the claimed "food effect" is an inherent property of oxymorphone itself, present both in controlled release and immediate release formulations of that drug. See In re Kubin, 561 F.3d 1351, 1357 (Fed. Cir. 2009) (stating "(e)ven if no prior art of record explicitly discusses the (limitation), (applicant's) application itself instructs that (the limitation) is not an additional requirement imposed by the claims on the (claimed invention), but rather a property necessarily present in (the claimed invention)"); see also King Pharmaceuticals, Inc. v. Eon Labs, Inc., 616 F.3d 1267, 1275-76 (Fed. Cir. 2010) (stating that "merely discovering and claiming a new benefit of an old process cannot render the process again patentable" (citations omitted)). This is not a case where the Board relied on an unknown property of prior art for a teaching. Rather, Maloney's express teachings render the claimed controlled release oxymorphone formulation obvious, and the claimed "food effect" adds nothing of patentable consequence.

[emphasis added]

In a petition for rehearing and en banc filed on June 23, Endo asserts that in dealing with Endo’s ‘859 application, the BPAI decision, and subsequently, the CAFC panel (Judges Linn, Moore and Rader) expanded the scope of the inherency doctrine to include not only anticipation but obviousness as well, in contrast to well-settled case law, the patent statute itself, and good public policy. 

In its petition, Endo argues that the BPAI and the CAFC erred, in that (a) the claim is not directed to a pharmaceutical composition per se, but to a method of treatment, (b) all parties agree that the “food effect” was not in the prior art and would not have been recognized by one of skill in the art prior to the applicants’ own disclosure of that effect, and therefore (c) in accordance with the CAFC’s own precedent, the claimed method cannot be deemed obvious, since inherency applies only with respect to novelty analyses, not obviousness analyses.

As the petition for rehearing notes, if the inherency doctrine is expanded to encompass obviousness as well, then

“The Panel's decision would bar claiming a new method of treatment using a known medicine (e.g., discovering that the proverbial cure for cancer is a new use of an old compound). Even if the use is novel and not what is expected by those of ordinary skill in the art at the time, the Panel decision would render it unpatentable because the effect would be inherent in the compound's use; although beneficial and previously unknown, the use would be deemed "inherent" to the compound.”

That’s clearly not a desirable result, and as the request notes, it’s a result that’s contrary to the constitutional mandate of promoting “the Progress of … useful Arts”.  

The petition also argues that under CAFC case law, obviousness, per 35 U.S.C. §103, is a question that must be decided with regard to the claimed subject matter as a whole, on the basis of what was understood by persons skilled in the art at the time the invention was made, and that prior to Endo’s own disclosure in the subject patent application, no one had taught or suggested that administering an oxymorphone sustained release formulation to a fed patient could yield significantly better results than administering the formulation to a fasting patient.  The request also explains why the cases relied upon by the CAFC in its decision do not support its position.

With regard to the application of the inherency doctrine, Endo’s arguments are compelling.  Suppose a compound is known for treating headaches; now it’s found that it can also be used as a chemotherapeutic agent.  But those anti-cancer properties are an inherent property of that compound, so under the CAFC’s logic in Kao, the fact that prior art taught administering the compound to a person, coupled with this inherent chemotherapeutic property, means that it was obvious to use the drug for chemotherapy – even though in point of fact it wasn’t obvious to anyone.  At any rate, apparently the USPTO thinks enough of this argument that it warrants further study – it has asked for an received an extension until August 12, 2011 to file its response to the petition for rehearing. 

Where I wonder about the propriety of Endo’s argument is not with regard to its observation about the proper scope of the inherency doctrine, but with regard to whether or not the doctrine has been misapplied with respect to the particular claims at issue, claims 8 and 21 of the ‘859 application.  That’s because although the “food effect” is recited in both claims, it’s open to interpretation if the recitation of the “food effect” constitutes a method step.  Suppose we left the “food effect” recitation out of claim 8.  Then it would read,

8.         A method for treating pain in a human subject in need of acute or chronic pain relief, comprising the steps of:

(a) providing a solid oral dosage form comprising about 5 mg to about 80 mg oxymorphone or a pharmaceutically acceptable salt thereof in a controlled release delivery system with a release rate profile designed to provide an adequate blood plasma level over at least 12 hours to provide sustained pain relief over this same period, the system comprising a filer and a hydrophilic material, wherein oxymorphone is the sole active ingredient; and,

(b) administering the dosage form to the subject.

In this instance, the argument for the obviousness of the claimed method on the basis of Maloney is much stronger, since Maloney teaches similar compositions using opioid analgesics (albeit not oxymorphone itself) and their administration to relieve pain.  If the clause “wherein the oxymorphone C_max is at least about 50% higher when the dosage form is administered to the subject under fed versus fasted conditions“ is viewed not as reciting a method step, but merely reciting what is an inherent property of the composition itself, then as the CAFC noted, that recitation doesn’t add patentable weight to the claim. 

Personally I’d feel that Endo was on better grounds had it recited the “food effect” limitation in part (a), as a functional feature along with the structural features of the composition being administered, for example,

8.         A method for treating pain in a human subject in need of acute or chronic pain relief, comprising the steps of:

(a) providing a solid oral dosage form comprising about 5 mg to about 80 mg oxymorphone or a pharmaceutically acceptable salt thereof in a controlled release delivery system with a release rate profile designed to provide an adequate blood plasma level over at least 12 hours to provide sustained pain relief over this same period, the system comprising a filer and a hydrophilic material, wherein oxymorphone is the sole active ingredient, and wherein the oxymorphone C_max is at least about 50% higher when the dosage form is administered to the subject under fed versus fasted conditions; and,

(b) administering the dosage form to the subject.

Here, since the fasted-vs.-fed release profile is recited as one of the limitations on the compositions that can be used in accordance with the claimed invention, it’s easier to assert that this limitation is part of the claimed method (though not a slam-dunk).  An even better route would have been to recite administering the composition to a patient within (or not within) a certain amount of time in relation to eating.  The prescribing information for OpanaER^® says that it should be administered “on an empty stomach, at least 1 hour prior to or 2 hours after eating”.  If such a recitation had been included in the claim 8, it clearly would have been part of the claimed method, not taught or suggested in the prior art, and, as Endo notes, therefore not subject to a finding of inherency.

Given the way the CAFC not once but twice bent over backward to construe the claims in Prometheus v Mayo as actually reciting a process, it’s possible that here too, if rehearing en banc is granted, the court will deem the last “wherein” clause of claim 8 to recite a process limitation.  In any event, I hope the CAFC takes this opportunity to correct itself and clarify the applicability of the inherency doctrine.  The way the panel decision is worded, Endo is right about this spelling the end of second medical use patents.

If You Add A Dependent Claim And No One’s There to Hear It, Does it Make Any Noise? – Part III: It Does, But Only if the Forest is in the U.S.A.

In December I posted about In re Tanaka, a case before the CAFC in which the question was whether or not an applicant’s desire to add a new dependent claim, but to make no other changes to his application, was sufficient to give him grounds to amend the application under the reissue statute, 35 U.S.C. §251: 

Whenever any patent is, through error without any deceptive intention, deemed wholly or partly inopera­tive or invalid, by reason of a defective specification or drawing, or by reason of the patentee claiming more or less than he had a right to claim in the patent, the Director shall, on the surrender of such patent and the payment of the fee required by law, reissue the patent for the invention disclosed in the original patent, and in accordance with a new and amended application, for the unexpired part of the term of the original patent. No new matter shall be introduced into the application for reissue.     * * * 

The Board of Patent Appeals and Interferences had ruled that since Tanaka did not assert a defective specification or drawing, his desired amendment did not give him the right to seek reissue of his patent under that prong of the statute, and that the case law held that failing to include a dependent claim did not constitute claiming more less than the patentee had a right to claim – after all, by definition a proposed dependent claim must fall within the scope of some already-existing claim, so the patentee could not be deemed to have claimed more or less than he was entitled to.  Thus the BPAI left in place a strange situation in which a person who was only guilty of leaving out a dependent claim had no recourse, but if the patentee had really screwed up big time, for example by mislabeling a figure, making a mis-statement in a claim, or gaining allowance of a claim that encompassed the prior art, then that patentee could seek reissue on those grounds – and at the same time introduce new dependent claims as well. 

Last week a split panel reversed the BPAI.  Citing to In re Handel – one of the decisions of the Court of Customs and Patent Appeals (CCPA, the predecessor of the CAFC) relied upon by the BPAI – Judge Linn, writing on behalf of himself and Judge Bryson, said that in that case, Judge Rich had explicitly said that adding new dependent claims was a way of correcting a defect in a patent: claiming less than the patentee had a right to claim, in the sense that the patentee included fewer claims than he had a right to include.  Thus, said Judge Rich in Handel, “[t]he narrower appealed claims are simply a hedge against possible invalidity of the original claims should the prior use be proved, which is a proper reason for asking that a reissue be granted.”  Judge Linn then stated that “While this court has since characterized that view [of Judge Rich] as dictum, it has not departed from it.” 

He then went on to show how allowing reissue applications solely for the purpose of incorporating new dependent claims was long-established practice.  Among other things, he noted that

“Claims of narrower scope can be useful to clarify the meaning of broader, independent claims under the doctrine of claim differentiation. Phillips v. AWH Corp., 415 F.3d 1303, 1314 (Fed. Cir. 2005). And dependent claims are also less vulnerable to validity attacks given their more narrow subject matter. Thus, the omission of a narrower claim from a patent can render a patent partly inoperative by failing to protect the disclosed invention to the full extent allowed by law.” 

Addressing a concern raised by Teva in an amicus brief, namely that “the public has an interest in preventing patentees from seeking reissue only to add narrower claims because such practice limits the public’s ability to rely on what is actually claimed in an issued patent”, the majority noted that 35 U.S.C. 252 gives courts equitable discretion to allow third parties who “prior to the grant of a reissue, made, pur­chased, offered to sell, or used within the United States, or imported into the United States, anything patented by the reissued patent, to continue the use of, to offer to sell, or to sell to others to be used, offered for sale, or sold, the specific thing so made, pur­chased, offered for sale, used, or imported unless the making, using, offering for sale, or selling of such thing infringes a valid claim of the reissued patent which was in the original patent.”  Thus, presumably, if an originally issued broad claim could be shown to be infringed but invalid, and a narrower, infringed and valid claim was only added during reissue, a company that in the interim began to infringe the originally-granted broad claim, or that even merely made “substantial preparations” to do so, could seek equitable relief to continue its activities.

Judge Dyk dissented, arguing that not only was there was no CCPA or CAFC precedent directly on point, but that there was Supreme Court precedent (Gage v Herring, 107 U.S. 640 (1883)) that precluded reissue applications solely to add new dependent claims.  The majority distinguish Gage on grounds that that decision was given under a very different reissue statute.

In the companion Israel case I discussed, Schwarz Pharma chose not to appeal the hearing officer’s ruling that a new dependent claim could not be added after an application had bee allowed, so overruling that decision will have to await another day.

If You Add A Dependent Claim And No One’s There to Hear It, Does it Make Any Noise? Israel PTO Rules That Application Cannot be Amended to Add Only Dependent Claims After Application Has Been Published – Part I

NOTE: This is the first part of a two-part post.  This part focuses on US practice; the second part will focus on Israel practice.

During ex parte patent practice in both the USA and Israel, applicants are free to amend their claims, so long as the amendments are supported by the specification.  But what are a patentee’s options for amending the claims after grant? 

In the USA, a patentee may seek a certificate of correction to rectify clerical or typographical errors (35 U.S.C. §254 and §255), but more substantive matters – like substantive patentability – may be dealt with through two types of proceedings, reexamination and reissue. 

Reexamination (35 U.S.C. §§301-307 and §§311-318) can be thought of as a means for correcting a particular type of error by the USPTO: the failure to find and consider certain prior art publications during the examination of a now-issued patent.  This is reflected in the fact that re-examination proceedings are predicated on the patentee or a third party presenting to the USPTO one or more printed prior art publications that raise a substantial new question of patentability regarding at least one of the claims of the patent.  Unlike prosecution of a regular application, reexamination can only result in the allowance or rejection of each claim; continuations and divisionals cannot be filed in a reexam. 

During reexamination, patentees may amend their claims.  Although patentees may not broaden the scope of the claims during reexamination, they may add new dependent or even independent claims.  Thus, for example, reexamination of US 6047319 resulted in all 17 originally-granted claims being upheld as allowable, and the addition of nine new claims, including a new independent claim 21 as well as claims 18-20 which depend from claim 1. 

Reissue, in contrast, can be thought of as a means to correct patentee error in an issued patent.  Unlike reexamination, only the patentee can request reissue.  As set forth in the first paragraph 35 U.S.C. §251, reissue may be sought for two reasons: a defective specification or drawing, or claims that are overly broad or overly narrow.

Whenever any patent is, through error without any deceptive intention, deemed wholly or partly inopera­tive or invalid, by reason of a defective specification or drawing, or by reason of the patentee claiming more or less than he had a right to claim in the patent…

As with reexamination, during reissue proceedings the applicant may amend its claims and add new ones.  However, in contrast to reexamination, once reissue proceedings have begun, the applicant may file continuations and divisionals.  And if the reissue application is filed within two years of the grant of the original patent, the patentee may seek to broaden his claims.  Broadening, in this context, means amending or filing a claim that is in some way broader than the originally granted claims, even if the amended or new claim is in other ways narrower than the granted claims.  In re Self, 671 F.2d 1344 (CCPA 1982).

An interesting question that is now before the CAFC is whether or not a patentee may seek reissuance of a patent when the only change sought is to add a new dependent claim.  According to a memo circulated to the examining corps by then-Deputy Commission John Love in late 2007, as long as one of the two threshold criteria are met – defective specification or drawings, or overly broad/narrow claims – USPTO policy is that the patentee is entitled to seek reissue under the statute.  But if the only the failure of the patent alleged by the patentee is the failure to include a dependent claim, that does not constitute “the patentee claiming more or less than he had a right to claim in the patent”, as per 35 U.S.C. §251, and thus the patentee is not entitled to seek reissue. 

The result of the PTO’s position is that in cases where there is an error in the specification to which the patentee can point, he can use this to gain entry to the reissue process; once there, he may add dependent claims, without narrowing his already-granted independent claims.  Exhibit A: the reissue of US 5945207, which resulted in Re 40005.  In that case, the only amendment to the claims that the patentee sought to introduce was the addition of new dependent claims.  Under the USPTO’s view, that alone would have been insufficient as a basis for reissue proceedings.  But the patentee also sought to correct the units for saturation magnetization recited in the specification, from “kA/m” to “kAm²/kg” (compare col. 3, line 17 of the original patent with col. 3, line 21 of the reissued patent).  That opened the door for the reissue process, enabling the patentee to add new dependent claims 8-14.

However, as set forth in a 2009 BPAI decision, Ex Parte Tanaka, in the case in which the patentee seeks only to introduce a new dependent claim, without amending the specification or otherwise amending any of the granted claims, the PTO will not allow the reissue to proceed.  In that decision, the BPAI explained why it believes that CCPA and CAFC case law interpreting §251, including Hewlett-Packard Co. v Bausch & Lomb, Inc., 882 F.2d 1556 (Fed. Cir. 1989) and In re Handel, 312 F.2d 943 (CCPA 1963) does not allow reissue applications solely for the purpose of introducing new dependent claims.  Inter alia, the Board was of the view that if failure to include dependent claims was an error correctable by reissue, “then virtually every patent could be reissued out of hand”.  Tanaka is now on appeal before the CAFC; oral arguments are scheduled for January 10, 2011. 

(Interestingly, Tanaka had originally attempted to file a broadening reissue in which only claim 1 would be broadened, but this was rejected as an attempt to recapture subject matter surrendered during prosecution – the portions of claim 1 Tanaka sought to delete had been relied upon during prosecution of the patent to distinguish over the prior art.  Since recapture cannot be the basis for a reissue, the application was rejected on this as well as prior art grounds.  Eventually Tanaka decided to leave the original seven claims alone and to file two new independent claims and several claims dependent therefrom.  These new claims were rejected in 2005 on prior art grounds, not on the grounds that the application failed to qualify as a reissue application under §251.  Later Tanaka amended some of the new claims, but after the Examiner maintained his prior art rejections, in March 2007 Tanaka cancelled all the new claims, save for new claim 16, which he amended to depend from originally granted claim 1.  The Examiner then rejected the application on the grounds that the amendment necessitated the filing of a new reissue oath, although he indicated that the new claim 16 was in principle allowable.  Tanaka then filed a new reissue oath, in which he listed the error in the patent being “the features recited in dependent claim 16 were not included in the original patent”.  It was at this stage, in October 2007, that the Examiner rejected the application as not being subject to reissue, because “the error specified in the oath filed 9/24/2007 is not an error correctible by reissue.  The applicant has not specified and error that broadens or narrows the scope of the claims of issued patent number 6093991.  The original claim 1 remains in the current reissue application, therefore the broadest scope of the patent remains the same.”  In an advisory action mailed in December 2007, the Examiner further explained that “Retaining the broad patent claim without amendment is an indication that the broad patent claims [sic] is not inoperative to cover the disclosed invention.”)

As a practical matter, it’s clear that there are several other ways Tanaka could have approached the situation and still managed to get his dependent claim allowed.  He could have kept a continuation application pending; claim 16 would apparently have been allowed, pursuant to the filing of a terminal disclaimer.  A riskier but viable approach would have been to find some prior art that raises a substantial new question of patentability and to request reexamination on that basis.  That would have opened the door to adding new claims, assuming that the claims were allowable over the prior art that served as the basis for the reexamination request.  And as mentioned above, he could have pointed to an error in the specification as the basis for reissue; that too would have opened the door for adding new claims. 

Another possibility might have been to amend one of the allowed claims in a way that formally narrowed the claim, but without detriment to the commercial value of that claim.  Such an approach was used in the reissue of US 6910617.  There, the patentee originally sought to only add dependent claims, and was rejected for the same reason Tanaka’s reissue application was rejected.  In response, the applicant amended the word “axial” in claim 1 to read “coaxial” (although he also found an error in the specification which also served as the basis for reissue).  This narrowing amendment brought the application within the ambit of the §251, enabling the patentee to add independent claims (although he could have added dependent claims as well) and leading to the issuance of Re 41101.

Editorializing for a moment, it seems strange that if a patent has no apparent flaws except for the failure to include some additional dependent claims, the patentee has no recourse to get those additional dependent claims introduced.  If, however, there are some flaws, e.g. the patentee made a mistake in the specification or drawings, or found some prior art that calls the validity of the claims into question, then that patentee is rewarded with an opportunity to file new dependent claims.