Patent Law Blog (Patently-O)

UK Patent Case Lowers Bar on Utility (Industrial Application)

Human Genome Sciences v. Eli Lilly (UK Supreme Court, 2 Nov 2011) [Decision] Case No. [2011] UKSC 51.

The Supreme Court of the United Kingdom (UKSC) began hearing cases in 2009 — taking on the role of court-of-last-resort formerly played by the House of Lords. The UKSC normally sits in five-member panels — here the panel consisted of Lords Hope, Walker, Neuberger, Clarke, and Collins. 

This decision focuses on EPC and UK version of the utility doctrine — the requirement that a patentable invention be "susceptible of industrial application". In a unanimous decision, the court determined that US utility doctrine creates an unduly high bar of patentability.  Thus, rather than requiring proof of specific, credible, and substantial utility at the time of filing, the UK court agreed that HGS's genetic sequence coding for Neutrokine-α was patentable even though there was no known use of the protein at the time the patent application was filing. The patent did not reveal how the protein "could be used to solve any particular problem" nor did it identify "any disease or condition which it could be used to diagnose or treat."  Yet, the UK court held that the industrial application requirement was met because the protein a member of a "TNF ligand superfamily" and all members of that family have been associated with important biologic activity.  "[A]ll known members of the TNF ligand family were expressed on T-cells and were able to co-stimulate T-cell proliferation, and therefore Neutrokine-α would be expected to have a similar function."  The UK Court of Appeals (Sir Robin Jacob) had previously held the patent invalid.

In his opinion, Lord Neuberger explicitly rejected the US cases of Brenner v Manson, 383 U.S. 519 (1966) and in re Fisher, 421 F 3d 1365 (2005) — finding that "there are obvious risks in relying on US jurisprudence when considering the precise nature of the requirements of Article 57 in relation to a claim for a patent for biological material under the EPC." 

There have been moves over the past fifty years (and more) to harmonise patent law across jurisdictions (the EPC and TRIPS - the Trade-Related Aspects of Intellectual Property Protection - being two important examples), and it is a laudable aim to seek to ensure that all aspects of the law of patents are identical throughout the world. However, the achievement of such an aim is plainly not currently practicable, and, although they have a great deal in common, there are significant and fairly fundamental differences (over and above the different words used in Articles 52 and 57 of the EPC and section 101 of 35 USC) between US patent law and the EPC (two notorious examples being the first to file rule in Europe, and file wrapper estoppel in the US).

Accordingly, particularly when it comes to a nice question such as the precise delineation of boundaries between patentability and unpatentability on the ground of industrial application, it would be unsurprising if the law was not identical under the two jurisdictions.

Instead of following US law, the panel instead latched onto the jurisprudence of the EPO — the body that also interprets the European Patent Convention (EPC).

Notes:

• The UK court of first instance (Kitchin J) found the patent invalid while the EPO Board had held the patent valid.
• The patent in question is European Patent 0,939,804
• IPKAT
• Of course, it is interesting that both HGS and Lilly are US-based companies.

Tags: Utility

Accessing Brand-Generic Settlement Data

FTC v. Cephalon (E.D. Pa. 2010)

Peter Loftus at the Wall Street Journal has written a short article titled “Drug Firms Want Patent Documents Kept Secret.”  At issue is a large cache of brand-generic settlement data held by the FTC and DOJ. 

In 2008, the FTC sued Cephalon alleging antitrust violations based on a set of reverse-payment settlements to generic manufacturers companies. The settlements meant that Cephalon could retain market exclusivity for its major drug Provigil until 2012.

The Medicare Prescription Drug, Improvement, and Modernization Act of 2003 (“MMA”) requires pharmaceutical companies to submit (to the FTC and DOJ) most major pharmaceutical patent settlements; brand-generic marketing or licensing agreements; and generic-generic agreements regarding the 180 day exclusivity.  Up to now (and according to law), the US government has kept those settlements secret except for (1) times when it challenges a settlement as anticompetitive and (2) aggregate settlement data released in FTC reports.

In its lawsuit, Cephalon has asked the District Court to compel disclosure of the underlying settlement information. According to the defendant, the FTC has repeatedly cited its own analysis of the settlement data, and Cephalon is requesting the source materials “in order to be in a position to respond to any use of the studies in motion practice and to be able to cross-examine experts or other witnesses relying upon them.”  In response, the FTC argued that its studies should be available to the court even if it does not reveal the underlying data because “reliance on extra-record empirical studies for … facts that have relevance to legal reasoning, is a well-established practice in federal courts.”  Of course, the problem here is that the FTC is both the plaintiff and the creator of the empirical study.  In addition to the FTC, a group of 35+ pharmaceutical companies also filed a brief — arguing that the disclosure would be highly prejudicial to their interests in keeping the information secret.   (The pharma brief may have been quite expensive to draft — it was signed by lawyers from 21 different major law firms).

The MMA includes some secrecy language preventing the government from disclosing the submissions “except as may be relevant to any administrative or judicial action or proceeding.”

• File Attachment: Cephalon's Motion to Compel
• File Attachment: Pharma Response
• File Attachment: FTC Response
• WSJ Article
• FTC 2010 Study on Pay-for-Delay Drug Settlements

In re Glatt Air Techniques, Inc.: Single Embodiment Commercial Success

By Jason Rantanen

In re Glatt Air Techniques, Inc. (Fed. Cir. 2011)
Panel: Newman, Prost (author), Moore

Glatt Air Techniques involves the reexamination of Patent No. 5,236,503, relating to an improvement to an apparatus known as a Wurster coater used to coat particles such as pharmaceutical ingredients.  The improvement comprises a shield means, such as a physical shield or air wall, positioned adjacent the spray nozzle that prevents particles from prematurely entering the stream of the coating spray before the spray pattern has fully developed, thus causing blockages in the apparatus.  During reexamination, the examiner rejected the claim at issue as obvious in light of the prior art, a determination that was upheld by the Board of Patent Appeals and Interferences.  In reaching its conclusion, the Board also rejected Glatt's commercial success evidence because it related only to a single embodiment - a physical shield - and the claim also covered the use of an air wall.  Thus, the Board concluded, the evidence was not commensurate in scope with the claim.

On appeal, the CAFC first found that the Board had failed to establish a prima facie case of obviousness.  It also disagreed with the Board's rejection of Glatt's evidence of commercial success, expanding on its past holding that "commensurate" does not mean "every" embodiment.   "The fact that Glatt’s commercial products only contain one type of shielding means does not make its commercial success evidence irrelevant. Under the PTO’s logic, there would never be commercial success evidence for a claim that covers more than one embodiment."  Slip Op. at 10.  Citing past precedent, the CAFC went on to state that "we have consistently held that a patent applicant “need not sell every conceivable embodiment of the claims in order to rely upon evidence of commercial success.” Id. (quoting In re DBC, 545 F.3d 1373, 1384 (Fed. Cir. 2008).  Thus, "commercial success evidence should be considered should be considered 'so long as what was sold was within the scope of the claims.'” Id.

Comment: Note that simply being commensurate with the scope of the claim does not necessarily satisfy the nexus requirement, as illustrated by the outcome of In re DBC.

Update: PharmaPatents Blog provides a detailed discussion of the CAFC's reversal of the Board's prima facie case of obviusness, asking whether there was anything else the applicant could have done to avoid a rejection by the examiner and Board.   

Prometheus Laboratories v. Mayo: The Broad Scope of Statutory Subject Matter

By Jason Rantanen

Prometheus Laboratories, Inc. v. Mayo Collaborative Services (Fed. Cir. 2010)
Panel: Rader, Lourie (author), Bryson

Last Friday, the Federal Circuit released its second noteworthy post-Bilski decision (the first being Research Corp v. Microsoft).  The opinion, Prometheus v. Mayo, issued following a grant-vacate-remand order from the Supreme Court instructing the CAFC to revisit its original decision in light of Bilski.  Despite this procedural posture, however, the new opinion is quite similar to the old, arriving at the same conclusion through essentially the same reasoning.   

Prometheus initially came to the Federal Circuit following a district court grant of summary judgment of invalidity under § 101 (lack of patentable subject matter).  After the CAFC reversed the ruling of invalidity using its "machine-or-transformation" test, Mayo sought review by the Supreme Court.  The Court granted certiorari, vacated the CAFC decision, and remanded for consideration in light of its Bilski opinion.  Earlier Patently-O commentary includes a summary of the original Federal Circuit opinion and a discussion of the remand.

The patents-in-suit claim a method for determining whether a patient has received a therapeutically efficacious amount of drugs such as 6-mercaptopurine ("6-MP") and azthiopurine ("AZA"), which are used to treat inflammatory bowel diseases but can produce toxic side effects.  In the human body, these drugs metabolize into 6-MP metabolites, including 6-methylmercaptopurine ("6-MMP") and 6-thioguanine ("6-TG").  By administering the drug, measuring the subject's levels of 6-MMP and 6-TG and comparing them to pre-determined levels, toxicity can be minimized and efficacy maximized. 

Claim 1 of Patent No. 6,355,623 is representative:

1. A method of optimizing therapeutic efficacy for treatment of an immune-mediated gastrointestinal disorder, comprising:
    (a) administering a drug providing 6-thioguanine to a subject having said im-mune-mediated gastrointestinal disorder; and
    (b) determining the level of 6-thioguanine in said subject having said immune-mediated gastrointestinal disorder,
    wherein the level of 6-thioguanine less than about 230 pmol per 8x108 red blood cells indicates a need to increase the amount of said drug subsequently administered to said subject and
    wherein the level of 6-thioguanine greater than about 400 pmol per 8x108 red blood cells indicates a need to decrease the amount of said drug subsequently administered to said subject.

Although similar in most respects, some claims of the second patent-in-suit, No. 6,680,302, dispense with the "administering" step.

The Patents Do Not Claim a Physical Phenomena
On remand, the CAFC again rejected Mayo's argument that the '623 and '302 patents claim a "natural phenomenon."  In seeking a judgment of invalidity under Section 101, Mayo contended (and the district court agreed) that the "administering" and "determining" steps are merely necessary data-gathering steps for the use of the correlations between 6-TG and 6-MMP and therapeutic efficacy or toxicity in patients.  Because these correlations are simply natural phenomena, Mayo reasoned, they were unpatentable.

As before, the Federal Circuit disagreed, noting that Bilski provides a broad - although not unlimited - scope for patent protection, and “an application of a law of nature or mathematical formula to a known structure or process may well be deserving of patent protection.”  Slip. Op. at 12, quoting Bilski, 130 S.Ct. at 3230.  Furthermore, the court stated, neither the Supreme Court's order to vacate and remand the original Prometheus decision nor Bilski dictates a wholly different analysis or different result.  

The crux of the CAFC's determination that the asserted claims recite a patent-eligible application of naturally occurring correlations between metabolite levels and efficacy or toxicity as opposed to the natural correlation itself rests on the specific treatment steps recited by the claims: the "administering" step and the "determining" step.  "The inventive nature of the claimed methods stems not from preemption of all use of these natural processes, but from the application of a natural phenomenon in a series of steps comprising particular methods of treatment." Slip. Op. at 15-16.

In support of its conclusion, the court reiterated its earlier determination that the treatment methods in Prometheus's patents satisfy the "machine-or-transformation test.  Although this is not the exclusive test, post-Bilski, it nevertheless provides important clues to subject matter patentabilty.  In applying the machine-or-transformation framework, the court specifically rejected Mayo's argument that the disputed claims simply claim natural correlations and the associated data-gathering steps, broadly stating that the asserted claims are "claims to methods of treatment, which are always transformative when one of a defined group of drugs is administered to the body to ameliorate the effects of an undesired condition."  Slip Op. at 17.  Even leaving out the administration step does not make the claims unpatentable, as the CAFC also found the "determining" step to be transformative because it involves "[s]ome form of manipulation, such as the high pressure liquid chromatography method specified in several of the asserted dependent claims or some other modification of the substances to be measured, [which] is necessary to extract the metabolites from a bodily sample and determine their concentration."  Id. at 18.

In reaching this conclusion, however, the court was forced to distinguish its earlier decision In re Grams, 888 F.2d 835 (Fed. Cir. 1989), which similarly claimed a process that involved "(1) performing a clinical test on individuals and (2) based on the data from that test, determining if an abnormality existed and determining possible causes of any abnormality by using an algorithm."  Id. at 20.   Unlike the claims in Grams - which the CAFC found unpatentable "because the tests were just to 'obtain data'" (Slip Op. at 20) - the claims of the Prometheus patents "require the performing of clinical tests on individuals that were transformative." 

Mental Steps
In addition to its overarching analysis of the subject matter issue, the opinion also includes an interesting discussion of the use of mental steps in patent claims.  Although the CAFC agreed that the final "wherein" clauses are mental steps, "A subsequent mental step does not, by itself, negate the transformative nature of prior steps. Thus, when viewed in the proper context, the final step of providing a warning based on the results of the prior steps does not detract from the patentability of Prometheus’s claimed methods as a whole." Slip Op. at 21. Because no claim in the Prometheus patents claims only mental steps, "contrary to Mayo’s assertions, a physician who only evaluates the result of the claimed methods, without carrying out the administering and/or determining steps that are present in all the claims, cannot infringe any claim that requires such steps." Id.

Additional Commentary
In addition to an extensive discussion about the decision in response to Dennis's post on Sunday, other sites commenting on the decision include:

• Patent Docs
• patents4life
• IP Watchdog
• Chris Holman's IP Blog
• Hal Weger of Foley has suggested that another grant of certiorari may be down the road, given the opinion's refusal to discuss a three-Justice dissent from the dismissal of certiorari in Lab. Corp., 548 U.S. 124, that was cited with approval by five Justices in two concurrences in Bilski.

Cancer Research Technology v. Barr Laboratories: Prosecution Laches and Inequitable Conduct

By Jason Rantanen

Cancer Research Technology Ltd. v. Barr Laboratories, Inc. (Fed. Cir. 2010)
Panel: Newman, Lourie (author), Prost (dissenting)

Although overshadowed by the en banc Federal Circuit arguments in TheraSense v. Becton Dickinson this morning, Cancer Research Technology v. Barr Laboratories may provide a preview of what the opinions in TheraSense could look like - although it doesn't necessarily indicate which view of inequitable conduct will ultimately prevail.  In Cancer Research, Judges Lourie and Newman reversed a district court finding of prosecution laches and inequitable conduct, while dissenting Judge Prost would have reached the opposite result.

The patent at issue (the '291 patent) involved a set of thirteen tetrazine derivatives that the original 1982 specification identified as possessing anticancer activities based on animal studies.  During the first nine years of prosecution, the examiner repeatedly rejected the claims due to lack of utility; rather than file a response to the office actions, the applicant instead filed continuation after continuation.  In 1991, Cancer Research obtained ownership of the patent application and shortly thereafter responded to the examiner's arguments.  The patent issued in 1993 and expires in 2014. 

Following Cancer Research's clinical testing, the FDA approved one of the compounds covered by the '291 patent (marketed as TEMODAR) for the treatment of one type of cancer in 1999 and a second in 2005.  In 2007, Barr Laboratories filed an Abbreviated New Drug Application ("ANDA") for a generic form of TEMODAR.  Cancer Research sued Barr for infringement four months later.  During the district court proceedings, the parties stipulated to validity and infringement and, after a bench trial, the court found that the patent was unenforceable due to prosecution laches and inequitable conduct. 

Prosecution Laches
Given the nearly ten-year delay before any meaningful response to the examiner's rejection was filed, Barr contended that the patent was unenforceable due to prosecution laches.  The district agreed, concluding that the delay in prosecution was unreasonable and unexplained.

On appeal, the majority reversed the finding of prosecution laches, holding that the doctrine requires not just unreasonable delay, but also a showing of prejudice.  The majority further held that "to establish prejudice, an accused infringer must show evidence of intervening rights, i.e., that either the accused infringer or others invested in, worked on, or used the claimed technology during the period of delay."  (Slip Op. at 9).  Here, there was no evidence of intervening rights during the prosecution period, such as evidence showing that someone other than the patent holder attempted to develop the claimed compounds.  Even Barr itself waited until 2007 - four years later than required - before filing its ANDA.  The majority also noted that there was no public harm: in the absence of the patent, Cancer Research likely would not have been incentivized to develop TEMODAR at all. 

Writing in dissent, Judge Prost rejected the notion that prosecution laches requires prejudice, let alone intervening rights; rather, under her reading of the precedent such a requirement is not part of the laches determination.  Furthermore, in her opinion, both Barr and the public were harmed by Barr's inability to market a generic version of TEMODOR.

Comment: I'm unconvinced by Judge Prost's argument on this point.  If she is correct, then the '291 patent was never enforceable - be it in 2007, when Barr filed its ANDA, or 1993, when it issued.  Yet without an enforceable patent, Cancer Research would never have developed TEMODOR, let alone engaged in the expensive Phase III clinical studies necessary to demonstrate its safety and efficacy.  Thus, the "harm" to the public would been greater in the absence of the '291 patent, not less.

Inequitable Conduct
The majority also rejected the district court's finding of inequitable conduct, while the dissent reached the opposite conclusion.  Both opinions focused on the subject of intent to deceive. 

The inequitable conduct issue in this case revolved around an extensive series of articles by an inventor on the '291 patent that presented data from post-application clinical trials of the claimed compounds.  These articles included conclusions indicating that some of the claimed compounds demonstrated high toxicity and low anticancer activity, which the district court found to be highly material to the patent claims because it directly contradicted statements in the '291 patent regarding the compounds' utility in treating cancers, as well as the patentability of a broadly written claim.

The majority, following the "intent cannot be inferred from materiality" line of thought, concluded that the district court's only basis for finding intent was its determination that the withheld articles were highly material.  "Because the district court did not rely on any other evidence to support its finding of deceptive intent beyond that used to find the withheld data material, the court in effect relied solely on its materiality finding to infer intent to deceive."  Slip Op. at 17.  The majority also concluded that the inference of deceptive drawn by the district court about the inventor's publication of data was not the only reasonable inference; rather, the broad publication of this data in multiple articles is inconsistent with an inference of intent to deceive.  Thus, an equally reasonable inference is that the inventor did not appreciate the potential importance of the published data to the patentability of the patent claims.

Judge Prost, again writing in dissent, would have affirmed the district court's determinations.  In contrast to the majority, which required independent evidence of intent, her view of inequitable conduct is that it does not require separate evidentiary bases for materiality and intent; rather it is appropriate to cite to the same evidence for materiality and intent.  Furthermore, here there was additional evidence of intent in the form of the district court's credibility findings, "which are virtually unreviewable by this court."  Thus, under her approach to intent in inequitable conduct, "[w]e should not draw inference that the district court has already excluded based on its own credibility findings."

Comment: The majority and dissent's views on intent can be partially reconciled under the position that, although an intent to deceive may be partially based on evidence of materiality, materiality cannot be the sole basis for the finding of intent to deceive.  Here, in the majority's opinion, the finding of materiality was the sole basis for the intent to deceive determination, because the only additional factor - the credibility determination - was based on an erroneous inference.  On the other hand, in the dissent's view credibility determinations are unreviewable and are sufficient to provide the "beyond the materiality" support for an intent to deceive finding.

AstraZeneca v Apotex: Affirmance of a Preliminary Injunction

By Jason Rantanen

AstraZeneca LP v. Apotex, Inc. (Fed. Cir. 2010)
Panel: Rader, Bryson (dissenting in part), Linn (majority author),

Contrary to popular opinion, it's still possible to obtain a preliminary injunction in a patent case - it's just very difficult.  Astrazeneca v. Apotex provides an example of an affirmed preliminary injunction, and is significant for that reason alone.  It also raises some interesting inducement issues relating to intent that I'll discuss in a separate post.

Background
This case revolved around AstraZeneca's budesonide inhalation suspension drug product, which consists of a vial containing a single dose of budesonide suspended in a sterile liquid.  The drug is administered by squeezing the entire contents of the vial into a jet neubulizer, then inhaling the resulting mist through a mask attached to the nebulizer.  Because budesonide is an inhaled corticosteroid, the FDA requires that the label include a warning instructing patients to "titrate down" to the lowest effective dose of the medication to avoid any adverse effects from excessive use of the medication. 

The two patents at issue, Nos. 6,598,603 and 6,899,099, both contain method claims covering the once-daily use of the nebulized dose of a budesonide composition and product claims covering AstraZeneca's drug product kit.

Apotex sought approval to market a generic version of AstraZeneca's drug product.  As an ANDA applicant, it was well aware of AstraZeneca's patents, and sought to avoid the once-daily method claims by removing any mention of once-daily dosing from its labels.  While it succeeded in part, the FDA nevertheless required Apotex to keep the titration warning language in the generic product's label.

During the preliminary injunction proceedings, Apotex raised two principal arguments in response to the method claims.  First, it contended that they were anticipated; second, it contended that its distribution of the generic version of the drug would not induce infringement of AstraZeneca's method claims.  The district court rejected Apotex's arguments, and granted a preliminary injunction enjoining Apotex from marketing its product.

Note: The district court agreed with Apotex that the kit claims were invalid.  On appeal, the panel affirmed that determination. 

Anticipation
Apotex's first anticipation argument, involving a prior art patent, turned on a question of claim construction: whether the term "budesonide composition" encompassed budesonide encapsulated in liposomes (the '603 patent teaches dispersing budesonide in a solvent, either as a solution or a suspension that may include liposomes as an excipient).  The majority agreed with AstraZeneca's position, focusing on the discussion contained within the specification and buttressing its conclusions with extrinsic evidence (in this case, expert testimony).  Judge Bryson, dissenting on this point, reached the opposite conclusion: the claim term is not limited to budesonide directly dispresed in solvent, and thus the method claims are anticipated.

Apotex's also argued that a prior art British advertisement for AstraZeneca's product describing it as a twice-daily product anticipated the patents.  The majority again agreed with AstraZeneca that this reference did not anticipate the once-daily method claims, both because it did not disclose once daily dosing and because it was not enabled with respect to that type of dosing.  (Obviousness was apparently not in dispute, as at the time of the earlier reference no one recognized that the product could be administered once per day and still be effective).  Judge Bryson again disagreed, reading the prior art advertisement to suggest the administration of the drug once a day.

Inducement of Infringement
In challenging the district court's finding of inducement of infringement, Apotex focused on the subject of intent, arguing that its instructions did not demonstrate intent to cause the users of its product to engage in once-daily dosing and that it lacked specific intent to cause infringement of Apotex's patent.  The Federal Circuit rejected these arguments, affirming the district court's conclusion that the downward-titration instructions would necessarily result in some users engaging in once-daily dosing and noting that Apotex was well aware of the infringement problems raised by once-daily dosing, yet chose to proceed nonetheless.

Preliminary Injunction "Substantial Question of Invalidity" Standard
Although not playing a major role in the ultimate outcome of this appeal, the court's articulation of the "likelihood of success" standard, along with the subsidiary "substantial question of invalidity" element, is noteworthy given the divergent views on this subject, such as those expressed by Judges Newman and Gajarsa in Abbott Laboratories v. Sandoz, 544 F.3d 1341 (Fed. Cir. 2008).  AstraZeneca v. Apotex follows the approach of requiring a seemingly high threshold for patentees/low threshold for defendants on this issue:

For a patentee to establish that it is likely to succeed on the merits, it “must demonstrate that it will likely prove infringement of one or more claims of the patents-in-suit, and that at least one of those same allegedly infringed claims will also likely withstand the validity challenges presented by the accused infringer.” Ama-zon.com, 239 F.3d at 1351. When reviewing the grant of a preliminary injunction, this court “views the matter in light of the burdens and presumptions that will inhere at trial.” Titan Tire Corp., 566 F.3d at 1376 (citation omitted). A preliminary injunction should not issue if an alleged infringer raises a substantial question regarding either infringement or validity, i.e., the alleged infringer asserts an infringement or invalidity defense that the patentee has not shown lacks substantial merit. Genentech, Inc. v. Novo Nordisk A/S, 108 F.3d 1361, 1364 (Fed. Cir. 1997).

Teva v. Eisai: Standing for subsequent Paragraph IV filers

By Jason Rantanen

Teva Pharmaceuticals USA, Inc. v. Eisai Co. (Fed. Cir. 2010)
Panel: Rader, Dyk, and Prost (author)

A company seeking to market a generic version of a previously approved pharmaceutical product must file and receive approval of an Abbreviated New Drug Application (ANDA).  One aspect of the ANDA, called a "Paragraph IV Certification," involves an assertion that each of the patents for that drug listed in the Orange Book is invalid and/or not infringed.  The first manufacturer to file a Paragraph IV Certification is entitled to 180 days of generic marketing exclusivity; until the first-filer's exclusivity period has run, the FDA may not approve ANDA applications by other manufacturers who have filed Paragraph IV certifications for the same patent.   The first-filer's exclusivity period can be triggered by either (1) its commercial marketing of the drug; or (2) entry of a court judgment finding the patent invalid or not infringed.

Eisai holds the New Drug Application for donepezil, and owns five patents listed for this product in the Orange Book.  Of these five patents, one (the '841 patent) was the subject of separate litigation that produced a preliminary injunction barring Teva from marketing a generic version of donepezil until the expiration of the '841 patent in November 2010.  With respect to the remaining four patents, Ranbaxy Laboratories was the first to file a Paragraph IV certification; thus, because Ranbaxy's exclusivity period had not yet run, approval of Teva's ANDA was barred. 

To remedy this issue, Teva brought a declaratory judgment action against Eisai on the remaining four patents, seeking to trigger the exclusivity period.  Ultimately, there was no question as to whether Teva infringed the patents: Eisai had filed a statutory disclaimer for two of them, effectively canceling their claims, and had entered into a covenant not to sue Teva on the remaining two patents.  Nevertheless, all four patents continued to be listed in the Orange Book, precluding approval of Teva's ANDA.

After the Declaratory Judgment action was filed, Eisai moved to dismiss on the ground that Teva had failed to  establish the existence of an Article III case or controversy.  The district court agreed, dismissing the case for lack of jurisdiction.  Teva appealed.

The Federal Circuit reversed the dismissal.  With respect to the jurisdictional issue, the court focused on two cases: Caraco Pharmaceutical Laboratories, Ltd. v. Forest Laboratories, Inc., 527 F.3d 1278 (Fed. Cir. 2008) and Janssen Pharmaceutica, N.V. v. Apotex, Inc., 540 F.3d 1353, 1359 (Fed. Cir. 2008).   Looking to Caraco, the court explained that:

the generic drug company’s injury (i.e., exclusion from the market) is fairly traceable to the defendant’s actions because “but-for” the defendant’s decision to list a patent in the Orange Book, FDA approval of the generic drug company’s ANDA would not have been independently delayed by that patent. 527 F.3d at 1292; see 21 U.S.C. § 355(j)(5)(B)(iv). When an Orange Book listing creates an “independent barrier” to entering the marketplace that cannot be overcome without a court judgment that the listed patent is invalid or not infringed—as for Paragraph IV filers—the company manufacturing the generic drug has been deprived of an economic opportunity to compete. Id. at 1293; see also 21 U.S.C. § 355(j)(5)(B)(4). A declaratory judgment redresses this alleged injury because it eliminates the potential for the corresponding listed patent to exclude the generic drug from the market.

Slip. Op. at 11.  On the other hand, as in Janssen, where a generic manufacturer agrees that an Orange Book patent is valid, enforceable, and infringed, that generic manufacturer lacks standing to pursue a declaratory judgment action against other Orange Book patents for that drug because it cannot market its product even if the exclusivity period runs.  

Applying this framework, the court concluded that Teva had established an actual controversy.  The court distinguished Janssen on the ground that the preliminary injunction associated with the '841 patent was necessarily preliminary, and there was no final determination as to the validity, infringement or enforceability of the '841 patent.  Thus, although Teva could not currently market its product, the preliminary injunciton did not eliminate the existence of a controversy.

With respect to the discretionary component of the Declaratory Judgment Act, the Federal Circuit concluded that the district court had abused its discretion in declining jurisdiction, both because it had relied on its determination that it lacked subject matter jurisdiction as well as being unsupported by the facts. 

Sun Pharmaceuticals v. Eli Lilly: obviousness-type double patenting in the pharmaceutical context

Sun Pharmaceutical Industries, Ltd. v. Eli Lilly and Company (Fed. Cir. July 28, 2010)

By Jason Rantanen

Double-patenting issues arise when two commonly owned applications cover the same or similar inventions.  The issues in this appeal revolved around an earlier patent claiming a composition of matter and describing a method for using that composition, and a later patent claiming that method of use. 

Both of the patents in this case, Patent No. 4,808,614 (the '614 patent) and Patent No. 5,464,826 (the '826 patent) relate to gemcitabine, the active ingredient of Lilly's Gemzar® product.  The '614 patent claims both gemcitabine itself, as well as a method of using it to treat viral infections.  In addition, the '614 patent's specification discloses using gemcitabine to treat cancer.  The '826 patent claims a method of treating cancer comprising administering a therapeutically effective amount of gemcitabine.  The difference was important: the '614 patent expired on May 15, 2010, while the '826 patent does not expire until November 7, 2012.

Note: The applications leading to both the '614 and '826 patents were filed on the same day, December 4, 1984.  The '614 was a continuation-in-part of application No. 473,883 ("the '883 application"), which did not disclose using gemcitabine to treat cancer.  That information was added as part of the continuation-in-part. 

After filing an Abbreviated New Drug Application ("ANDA") for a generic version of Gemzar®, Sun Pharmaceuticals, sought a declaratory ruling that the '826 patent was invalid and not infringed.  Lilly counterclaimed for infringement of the '826 and '614 patents.  The '614 patent was not at issue in this appeal.

Obviousness-type double patenting applies
Applicants are barred from obtaining multiple patents covering the same invention by the doctrine of double patenting.  There are two types of double patenting: statutory double patenting, which prohibits a later patent from covering the identical invention, and obviousness-type double patenting, which prevents a later patent from covering a slight variation of an earlier patented invention.

On appeal, the panel agreed with the district court and Sun that the latter type of double patenting occurred here, thus invalidating the asserted claims of the '826.  The basis for the court's decision were two prior opinions, Geneva v. GlaxoSmithKline, 349 F.3d 1373, and Pfizer v. Teva, 518 F.3d 1353.  In Geneva, the earlier patent claimed a compound and the specification disclosed its effectiveness for inhibiting beta-lactamase.  The later patent claimed a method of using the compound to affect beta-lactamase inhibition.  Similarly, in Pfizer, the earlier patent claimed several compounds and the specification disclosed their use in treating inflamation; the later patent claimed a method of using these compounds for treating inflammation.  In both cases, the court ruled that the claims were not "patentably distinct," and thus the latter claims were invalid for obviousness-type double patenting.  

While Lilly argued that Geneva and Pfizer did not apply because "the specification of the earlier patent disclosed a single use for the claimed compound, which was an essential part of the patented invention and thus necessary to patentability," Slip Op. at 8, the court rejected that argument for two reasons.  First, the court disagreed that the specification in Pfizer disclosed more than one utility for the claimed compound.  In addition, the court read the rule of Pfizer as simply that "obviousness-type double patenting encompasses any use for a compound that is disclosed in the specification of an earlier patent claiming the compound and is later claimed as a method of using that compound.  Pfizer never implies that its reasoning depends in any way on the number of uses disclosed in the specification of the earlier patent."  Slip Op. at 10. 

The court also rejected Lilly's argument that the specification of an earlier application should have been consulted, as opposed to the specification of the '614 patent.  Drawing upon its claim construction precedent, the court noted that the specification is relevant to determining the coverage of the claims, which is at the heart of the obviousness-type double patenting analysis.  The court further noted that "consulting the specification of the issued patent, as opposed to an earlier version, is consistent with the policy behind double patenting," which rests "on the fact that a patent has been issued and later issuances of a second patent will continue protection, beyond the date of expiration of the first patent of the same invention or an obvious variation thereof."  Slip Op. at 14-15.

Federal Circuit Review of Patent Term Extensions

By Jason Rantanen

During the summer, the Federal Circuit is a relatively quiet place.  The judges often take their non-sitting months during this time, and the pace of opinions tends to drop.  Thus, for the next few weeks, I'll mostly be posting summaries of cases that issued this past spring and early summer.  The two cases discussed below deal with a relatively minor — but still important — issue in the pharmaceutical and medical device context: patent term extensions based on extensive regulatory review periods. 

Ortho-McNeil Pharmaceutical, Inc. v. Lupin Pharmaceuticals, Inc. (Fed. Cir.  May 10, 2010)
Photocure ASA v. Kappos (Fed. Cir. May 10, 2010)

35 U.S.C. §156 allows a patentee to obtains a term extension if the patent covers a product that has been subject to a regulatory review period before it can be marketed or used.  Pharmaceuticals and medical devices are subject to such a review period, and new drug products in particular often involve a lengthy application and testing process.  The initial determination as to whether a patent term extension should be granted is made by the USPTO, in consultation with the FDA.  That decision is subject to review or challenge in district court proceedings.

One of the key issues in determining whether a patent term extension is warranted for a drug is whether it is the first time regulatory approval has been granted for this particular drug product, a determination that turns on whether or not the "active ingredient" had previously been approved by the FDA.  Ortho-McNeil and Photocure, both authored by Judge Newman and issued on the same day, provide an interesting contrast on this issue. 

Ortho-McNeil v. Lupin
In Ortho-McNeil, the extension issue arose in the context of an injunction entered against Lupin Pharmaceuticals prohibiting it from making, using, selling, etc. a drug product covered by U.S. Patent No. 5,053,407 (the '407 patent) during the extension period.  In that case, the district court affirmed the PTO's determination that an enantiomer was a different drug product then its racemate. In doing so, the district court noted that the PTO's determination should be afforded great deference.

Note: Enantiomers are molecules that are mirror images of one another.  Due to their different orientation, they have different properties.  A racemate is a composition consisting of equal parts of the two enantiomers.  The '407 patent covered a substantially purified form of one of the two enantiomers (levofloxacin) in the racemate ofloxacin.  There was no dispute that levofloxacin was separately patentable from ofloxacin.

On appeal, the Federal Circuit agreed with the district court, concluding that there was no basis for challenging the established FDA and PTO practices of treating enantiomers as different drug products and rejecting Lupin's legislative intent argument.

Photocure v. Kapos
Photocure involved a contrary determination by the PTO: that the drug product at issue was not a different "active ingredient," and thus the patentee was not entitled to an extension.  In Photocure, the product at issue ("MAL") was a methyl ester of a compound ("ALA") that had previously been approved for the same therapeutic use.  While the FDA treated MAL as a new drug, requiring a full approval process, the PTO rejected the extension based on its conclusion that § 156(f)(2) does not mean the product approved by the FDA, but rather the "active moiety," which it concluded was the same in both MAL and ALA.

Both the district court and Federal Circuit disagreed.  In rejecting the PTO's interpretation of 156(f)(2), the Federal Circuit reasoned that § 156 focuses on the product that is subject to approval by the FDA, not the underlying pharmacological mechanism. Furthermore, Skidmore and Chevron deference standards did not apply because the statute was not unambiguous and the PTO's interpretation was neither persuasive nor consistent. 

Note: although not the primary focus of the opinon, the panel also concluded that the PTO was wrong even under its  "active moiety" interpretation as the biological properties of ALA and MAL are indisputably different.

* * * *

In addition to the issues discussed above, the scope of the injunction in Ortho-McNeil is worth noting.  Although the extension authorized by 35 U.S.C. § 156 covers the "selling" or "using" of the product covered by the patent, the district court enjoined Lupin from engaging in any of the traditional forms of direct infringement, including "making" or "importing."  Despite the literal language of §156, the Federal Circuit affirmed the scope of this injunction because there are no non-pharmaceutical "uses" of the drug product, a point that Lupin apparently conceded.  Although as a practical matter this distinction may be of little value, as pharmaceutical companies often have production facilities located outside the United States, it is something to consider when seeking or opposing litigation under § 156. 

Federal Circuit Orders Another Case Transferred Out of Texas

In re Hoffamann-La Roche

(Fed. Cir. 2009)(on writ of mandamus)

Novartis sued Roche and its partners in the Eastern District of Texas for infringement of its HIV treatment patent. After being denied by District Court Judge Folsom, Roche petitioned the Federal Circuit for a writ of Mandamus -- asking the appellate court to order the case to be transferred to the Eastern District of North Carolina. Following its own TS Tech precedent as well as the 5th Circuit's Volkswagen case, the Federal Circuit complied and has ordered the case transferred.

Under TS Tech, the appellate court will order a transfer on mandamus when the alternate forum is "clearly more convenient." Here, the appellate court found that important sources of proof are found in North Carolina (the location where the accused drug was developed); that the trial would impact the ongoing reputation of North Carolina residents (Duke professors); and that several non-party witnesses would be within the subpoena power of the North Carolina court. At the same time, the appellate court saw "no connection between this case and the Eastern District of Texas except that in anticipation of this litigation, Novartis' counsel in California converted into electronic format 75,000 pages of documents . . . and transferred them to the offices of its litigation counsel in Texas." Although not completely disregarding these litigation-preparation activities, the appellate court clearly deemed them less important in the venue considerations.

Irreparable Harm of Generic Competition: Federal Circuit Affirms Finding that Generic Entry Does not Cause Irreparable Harm

Altana Pharma & Wyeth v. Teva (Fed. Cir. 2009)

Altana's Patent No. 4,758,579 claims the proton pump inhibitor pantoprazole - the active ingredient the anti-ulcer drug Protonix®. Of course, PPI's were known before Altana's patent and even one of Altana's own prior patents discusses a "compound 12" that is structurally similar to those claimed in the '579 patent.

Teva and Sun filed for permission to begin making generic versions of the drug, and Altana subsequently filed this infringement action. (Altana filed separate actions that were consolidated.)

This appeal stems from the New Jersey district court's denial of Altana's motion for a preliminary injunction. The lower court found that the patentee had failed to prove two critical prerequisites of equitable preliminary relief: (1) a likelihood of success on the merits and (2) irreparable harm.

The equitable test for preliminary injunctive relief requires that the requesting party prove:

1. a reasonable likelihood of success on the merits;
2. irreparable harm if an injunction is not granted;
3. a balance of hardships tipping in its favor; and
4. the injunction’s favorable impact on the public interest.

Although the ultimate grant or denial of preliminary relief is within the "sound discretion of the district court," failure to abide by these four factors would be reversible error. Orders to grant or deny a preliminary injunction are immediately appealable.

Likelihood of Success: The Federal Circuit has held that preliminary relief should be denied if the accused infringer raises a "substantial question" of invalidity of the asserted claims. At the PI stage, the court need not consider the ultimate "clear and convincing" standard. Rather, the focus is on "vulnerability."

Obviousness of Chemical Compound: When considering the obviousness of a chemical compound, courts ordinarily first look for a "lead compound" known in the prior art and then consider whether a chemist would have had some reason to modify the known compound in the particular manner to achieve the new compound. Courts are not rigidly bound by this ordinary approach - thus, for instance, a court may look to multiple lead compounds:

Moreover, to the extent Altana suggests that the prior art must point to only a single lead compound for further development efforts, that restrictive view of the lead compound test would present a rigid test similar to the teaching-suggestion-motivation test that the Supreme Court explicitly rejected in KSR.

Here, the appellate panel found "ample evidence" that a chemist would have chosen "compound 12" as a natural choice for further PPI research. The particular modification of compound 12 was then suggested in articles by Sachs and Bryson who were researching properties of effective PPIs.

Considering this evidence as a whole, the Federal Circuit found it sufficient to raise a substantial question of obviousness.

Irreparable Harm: The district court could not find any irreparable harm of allowing infringement during the course of the litigation. Often, money damages are seen as insufficient when the defendant does not have cash-on-hand. Here, however, Teva and Sun both have plenty. The court also found that Altana almost certainly has a business plan to deal with the launch of generics. During the litigation, Nycomed purchased Altana -- seemingly in the lower court's view, that purchase also indicates that money damages are adequate (since a price can be placed on the company & its patent rights).

Perhaps most harmful to Altana was that the lower court found the patentee's statements of harms "exaggerated" and lacking "credibility." A court sitting in equity righty places a dim light on activities suggestive of unclean hands.

On appeal, the Federal Circuit affirmed without significantly commenting on the merits. Rather, the court made this case about equitable discretion: "the law cited by the district court highlights this court’s deference to a district court’s determination whether a movant has sufficiently shown irreparable harm."

Denial of Preliminary Injunction Affirmed

Judge Newman wrote a short concurring opinion.

Although the evidence presented to the district court does not, in my view, establish invalidity of the patent on the pharmaceutical product pantoprazole, see, Gonzales v. O Centro Espirita Beneficente Uniao do Vegetal, 546 U.S. 418, 429 (2006) ("the burdens at the preliminary injunction stage track the burdens at trial.") at this preliminary stage deference is warranted to the district court’s weighing of the conflicting expert opinions interpreting the evidence. On this basis, I concur in sustaining this discretionary action.

Federal Circuit Affirms District Court’s Extension of 30-Month FDA Stay

Eli Lilly v. Teva Pharmaceuticals (Fed. Cir. 2009)

Teva is hoping to make a generic version of Lilly's Evista brand raloxifene tablets that are used to help prevent postmenopausal osteoporosis. In May 2006, Teva filed an abbreviated new drug agreement (ANDA) and Lilly subsequently sued for patent infringement and to block the generic release. Under the law, after the patentee files suit, the FDA cannot then approve the generic for thirty months "unless the court has extended or reduced the period because of a failure of either [party] to cooperate reasonably in expediting the action." This is commonly known as the "thirty month stay."

In this case, the court originally set a trial date four months after the end of the thirty month period. In the months leading up to trial Teva altered its proposed generic formulation, which changed Lilly's litigation strategy. As a result, the district court ordered that the FDA thirty month stay be extended for four extra months – until May 2009. Teva filed an emergency appeal to lift the stay.

On appeal, a split Federal Circuit panel affirmed – finding that the district court acted within its discretion in extending the stay based on Teva's activity. "Trial courts, thus, may shorten or extend the thirty-month statutory period based on the parties' uncooperative discovery practices before the court." In its decision, the court distinguished the 2002 Andrx v. Bioval case. In Andrx, the Federal Circuit found that the district court had abused its discretion in shortening a thirty-month stay based on a party's "positions before the FDA." Rather, changes to the thirty-month period must be based on failure to cooperate court.

In dissent, Judge Prost argued that the majority misinterpreted the statute to grant too much deference to the district court in extending the stay. Rather, the statute requires that the stay should only be extended when a party fails "to cooperate reasonably in expediting the action." Here, Judge Prost argues, the lower court did not find that Teva failed to cooperate, but only that Lilly could use more time to respond. The statute is limited in a way that does not allow extension of the stay in that situation.

Notes:

• Judge Rader wrote the majority opinion and was joined by Chief Judge Michel.
• On the expedited schedule, Appellant filed its principle brief November 24, 2008; briefing was complete on December 30, 2008; Oral arguments were heard on January 14, 2009; and a decision rendered on February 24, 2009.

Patently-O Bits and Bytes No. 85

• BPAI Plummeting Reversal Rate. [LINK]
• Accessing Patently-O: I have set up several ways to access Patently-O depending upon your preferences:
  • Via the Website;
  • Via daily Email Digest;
  • Via Full Content RSS Feed;
  • Via RSS Squib.
• Obama's original selection for Secretary of Commerce – Bill Richardson – will not pan out. Richardson has decided to withdraw his nomination. Although we rarely see the Secretary of Commerce directly shaping USPTO policy, the Office is still part of the Department of Commerce and the USPTO Director reports directly to the Secretary of Commerce. The greatest impact of the Secretary of Commerce may be on the selection of high-level PTO officials (including the Director).
• The Patent Troll Tracker debacle made the ABA Journal's list of the top ten legal stories of 2008. (The only IP related issue to make the list). [Link]
• Patently-O Jobs:
  • Microsoft (Redmond, WA) needs a patent portfolio manager (at least 5 years exp.) [LINK]
  • Google (Mountain View, CA) needs an IP litigation counsel (at least 3 years exp.) [LINK]
  • Amin Hallihan (Chicago) focuses on IP & FDA issues. The firm is looking one or more experienced patent attorneys to join their ranks. [LINK]
• Phama Law: Pozen's migraine drug (Treximet) is special because it combines two types of migraine treatments into one pill: naproxen sodium (Aleve) and sumatriptan (Imitrex). Alphapharm (Mylan) filed an abbreviated new drug application (ANDA) to begin marketing a generic version of the drug – arguing that the patent is invalid. On January 2, Pozen sued Alphapharm for infringement in the Eastern District of Texas. Claim 1 of the asserted patent reads as follows:
  • 1. In a method for treating a migraine patient by administering a 5-HT agonist [Imitrex], the improvement which comprises: concomitantly administering to said patient a long-acting, non-steroidal, anti-inflammatory drug (LA-NSAID) [Aleve] in an amount that, together with said 5-HT agonist, is effective to reduce migraine relapse or produce longer lasting efficacy compared to the administration of said 5-HT agonist in the absence of said LA-NSAID.

Prior Art Must Enable a Skilled Artisan to Make the Invention without Undue Experimentation

Impax Labs v. Aventis Pharmaceuticals (Fed. Cir. 2008)

This appeal focuses on the question of when a prior art disclosure is sufficiently enabled.

An Aventis patent covers the use of RILUTEK (riluzole) to treat ALS. Impax filed an ANDA with the FDA (seeking to market a generic version of the drug). In subsequent litigation, Impax also alleged, inter alia, that the listed patent should be held invalid. In particular, Impax argued that an earlier Aventis patent (the '940 patent) anticipates the asserted Aventis patent (the '814 patent) by suggesting a class of compounds may be used to treat ALS. The district court, however found that the '940 patent "does not enable a person of ordinary skill in the art to treat ALS with riluzole and therefore does not anticipate claims 1-5 of the '814 patent."

Burden particularly heavy: A defendant who hopes to use previously considered art to invalidate a patent has a "particularly heavy" burden.

Thus, a party challenging patent validity has the burden to prove its case with clear and convincing evidence. When the examiner considered the asserted prior art and basis for the validity challenge during patent prosecution, that burden becomes particularly heavy. See Hewlett-Packard Co. v. Bausch & Lomb Inc., 909 F.2d 1464, 1467 (Fed. Cir. 1990).

In this case, the '940 patent had been considered by the examiner during prosecution of the asserted patent.

Enabled Prior Art: To anticipate, the prior art must be enabling – i.e., it must "enable one of ordinary skill in the art to make the invention without undue experimentation."  This enablement standard is different from the applicant's 112 enablement requirement which requires enablement of both making and using the invention.  That distinction becomes lost in cases like this, however, where the patent covers a method of treatment.  Here, the Federal Circuit appears to require anticipatory prior art to enable practicing the claimed method.

Enablement of prior art is a question of law, but is based on underlying factual findings. In close cases, the important factual finding is the amount of experimentation that would have been necessary. Applying Wands factors, the Federal Circuit agreed that the prior reference was not enabling.

As shown by the trial court, the [prior art] '940 patent's dosage guidelines are broad and general without sufficient direction or guidance to prescribe a treatment regimen. The alleged prior art also contains no working examples. Finally, nothing in the '940 patent would have led one of skill in the art to identify riluzole as a treatment for ALS. In sum, each component of the claimed invention—identifying riluzole as a treatment for ALS and devising dosage parameters—would require undue experimentation based on the teachings of the '940 patent.

Burden of Proving Enabled Prior Art: In judging anticipation, courts generally presume that prior art is enabled. That presumption may be overcome by a patentee by providing "persuasive evidence" of nonenablement. At that point, the ultimate burden of proof (with clear and convincing evidence) lies with the accused infringer. In this case, the district court did not explicitly follow this burden shifting framework. On appeal, the Federal Circuit held that articulation of the rule was not necessary. The burden was properly shifted because Aventis presented "sufficient evidence to overcome the presumption of enablement."

Holding: Affirmed. Validity challenge defeated because prior art was not enabled.

Notes:

• The opinion did not focus on timing: As technology develops more and more prior art references become enabled. Thus, it is important to consider at what point in time the reference must be enabling.

Disclosure: My former colleagues at MBHB represent Aventis in this case. The district court case was ongoing while I was there.

Appellate Court Affirms that Generic Omeprazole does not Infringe Prilosec Patent

IN RE OMEPRAZOLE PATENT LITIGATION (Fed. Cir. 2008) (nonprecedential)

Omeprazole is the active ingredient in the best-selling drug Prilosec. Mylan and others challenged Astrazeneca’s patents on grounds that their generic formulations do not infringe two Astra’s listed patents. “After a forty-two day bench trial,” a Southern District of New York district court agreed that the generic formulation do not infringe. On appeal was the question of whether the generic versions contained an “alkaline reacting compound” (ARC) as required by the claims.

Using the Specification as Proof of Non-Infringement: Astra argued that the talc used by the generic products included an ARC.  The CAFC agreed that the lower court had properly rejected that argument based in part on language in the specification. Specifically, the nearly identical specifications listed several different ARCs but did not include talc on the list.  “In contrast, the specifications also lists a number of ordinary excipients, among which is talc. . . Thus, the specifications themselves indicate that ARCs do not include talc.”

Defying conventional wisdom at the time, Mylan launched its generic version of Omeprazole in 2003 — despite ongoing patent litigation.

Non-infringement affirmed.

KSR Extended to Obviate Component Purified from Known Mixture

Aventis Pharma & King Pharma v. Lupin Ltd. (Fed. Cir. 2007).

Altace is the King/Aventis brand of ramipril – a top-selling ACE inhibitor.  The patent claims ramipril formulated “substantially free of other isomers.” The district court found the patent not invalid, but only by a slim margin. On appeal the CAFC reversed – finding the patent invalid as obvious.

Prior Art: One reference (‘944 patent) was filed as a continuation-in-part of an already abandoned patent application. That application was eventually revived, but Aventis argued on appeal that the ‘944 patent should not be awarded the filing date of the parent.  The CAFC found this a potentially interesting argument, but refused to hear the argument because Aventis had failed to make the argument at the district court level.

The appellate panel also agreed that the experimental results of a Shering Doctor constituted 102(g)/103(a) prior art because the Doctor had not abandoned, suppressed, or concealed her prior invention.

Obviousness of a Purified Form: The district court, ruling pre-KSR, found the patent nonobvious. On appeal, the CAFC took a different view – finding that the purified form of a known mixture is prima facie obvious if a PHOSITA would have some reason to believe that the mixture derives properties from particular components.

However, if it is known that some desirable property of a mixture derives in whole or in part from a particular one of its components, or if the prior art would provide a person of ordinary skill in the art with reason to believe that this is so, the purified compound is prima facie obvious over the mixture even without an explicit teaching that the ingredient should be concentrated or purified.

The prima facie case of obviousness is especially difficult to rebut where, as here, the potency of the mixture varies directly with the amount of isomer in the mixture.

The court implicitly distinguished this case from Forest Labs — noting that obviousness may be rebutted by showing difficulty in purifying the mixture.

[A] purified compound is not always prima facie obvious over the mixture; for example, it may not be known that the purified compound is present in or an active ingredient of the mixture, or the state of the art may be such that discovering how to perform the purification is an invention of patentable

Reversed, Patent Invalid as Obvious

Zyprexa Patent Upheld on Appeal

Eli Lilly v. Zenith Goldline (Fed. Cir. 2006)

Lilly’s patent covers olanzapine (Zyprexa®) and its use for the treatment of schizophrenia. Three generic manufacturers (Zenith (IVAX), Dr. Reddy’s, and Teva) filed an ANDA and Lilly responded with a complaint in the Southern District of Indiana.  After a bench trial, the district court agreed with Lilly that the patent was valid, infringed, and enforceable.  The defendants appealed to the Court of Appeals of the Federal Circuit.

Anticipation: To anticipate the invention, a prior art reference “must disclose each and every feature of the claimed invention, either explicitly or inherently.”  However, in both Petering and Schaumann, prior art references that disclosed the family of the claimed compound were found to anticipate the claimed compound — even though the claimed compounds were not specifically discussed.

Here, the CAFC found that those cases were not applicable because the closest reference to olanzapine did not spell out “a definite and limited class of compounds that enabled a person of ordinary skill in the art to at once envisage each member in this limited class.”

Obviousness: The CAFC agreed that the prior art references did not suggest a the compound. In addition, Lilly provided strong evidence of secondary considerations, including: “(1) a long-felt and unmet need; (2) failure of others; (3) industry acclaim; and (4) unexpected results.”

The record shows a long-felt need for a safer, less toxic, and more effective clozapine-like drug; a decade (or more) of failure to find a replacement for clozapine; a reasonable amount of commercial success for olanzapine; and a number of awards for olanzapine as indicators of industry acclaim.

Public Use: Prior to filing the patent application, Lilly conducted Phase I clinical safety trials. The court found, however, that the trials were well within the experimental use exception:

In this case, Lilly tailored its tests to their experimental drug safety and efficacy purpose, adequately monitored for results, and maintained confidentiality throughout the duration of the study. The trial court did not err in finding no public use.

 

In Preliminary Injunction Decision, CAFC Forecasts Post-eBay Jurisprudence

Sanofi v. Apotex (Fed. Cir. 2006)

Generic manufacturer Apotex wants to make something similar to Sanofi’s Plavix (clopidogrel bisulfate) and filed an ANDA alleging that Sanofi’s patents were invalid.

The two parties worked together to negotiate a settlement.  The agreement, however, was not accepted by state attorneys general even after new terms were presented. Under provisions of the agreement, the regulatory denial killed the settlement and the parties resumed litigation.

Sanofi then filed a motion for preliminary injunction to stop Apotex from selling its product. Within 21 days, the district court issued a PI. (During that time, Apotex shipped six-months of product). Apotex then appealed the PI.  Preliminary injunction jurisprudence has its own four-factor test that is similar to that of permanent injunctions. The plaintiff must show:

• Reasonable likelihood of success on the merits of the case;
• Irreparable harm if an injunction is not issued;
• Balance of hardships tipped in favor of the plaintiff; and
• Public interest that is not negatively impacted.

The major difference between the factors for consideration in preliminary injunctive relief and those for permanent relief is that preliminary relief requires a showing of a likelihood of success while permanent relief requires success on the merits as a precondition. Thus, the final three factors will give some indication of how the court will rule in post-eBay injunction cases.

Likelihood of success on the merits: Apotex took the odd position of arguing anticipation based on a broadly worded claim of a prior art patent that was examined during prosecution.  The CAFC confirmed that this made the burden of proving invalidity at trial “especially difficult.”  On obviousness, the CAFC confirmed that the unpredictability of enantiomer activity made the claimed dextrorotatory formation nonobvious even if the molecule as a whole was known.

On Irreparable Harm: The settlement agreement between the parties included a provision that capped any damages for infringement by Apotex — seemingly an admission that Sanofi would settle for money damages. The court did not buy-into this argument, but only because of the technicality that the agreement also contemplated an injunction.

Balance of hardships tip entirely in Sanofi’s favor because Apotex chose to launch its product under threat of injunctive relief.  It could have avoided the situation altogether and thus should not benefit from this factor.

Public Interest: The CAFC continued its line of the “importance of the patent system in encouraging innovation.” Interestingly, the court focuses on how the expense of pharmaceutical inventions necessitates strong patent protection. . . . begging the question of whether less expensive innovations (such as software) have less of a public interest in strong patent protection.