Federal Circuit Upholds Patent Infringement Ruling Against Microsoft

The Federal Circuit on Tuesday affirmed [.pdf] a lower court ruling that ordered Microsoft to stop selling its ubiquitous Word program and fined the company $290 million.  A Texas jury previously found that Microsoft’s Word 2007 infringed a software patent held by a Toronto-based technology company, i4i Inc.  Following that jury verdict, U.S. District Judge Leonard Davis issued a permanent injunction that prohibited Microsoft from selling the 2003 and 2007 versions of Word, and also ordered the company to pay more than $290 million in penalties.

The infringing versions of Word employ a technology called “Custom XML,” which lets users edit XML, the programming language that controls the way programs interpret and display the contents of a document.

In a statement, Microsoft said that it has been developing modified versions of Word 2007 that do not use the Custom XML feature.  Microsoft claims that these modified versions will be ready for sale by January 11, 2010, the date on which the permanent injunction takes effect.  Microsoft also noted that it is considering further legal options, “which could include a request for a rehearing by the Federal Circuit Court of Appeals en banc or a request for a writ of certiorari from the U.S. Supreme Court.”

The court noted that its decision did not impact copies of Word 2007 sold before the injunction takes effect, so Microsoft can continue to provide technical support with respect to those versions.

Pfizer Buys Wyeth, Industry Consolidates, IP Lawyers Rejoice

Pfizer’s purchase of Wyeth is big news for the pharmaceutical industry, and for big business in general.  Because its blockbuster cholesterol drug Lipitor comes off patent in 2010, probably ceding much of its $12.6 billion market to generics, Pfizer went looking to boost its pipeline in search of the next blockbuster drug.

Despite the generally gloomy outlook on Pfizer’s own pipeline, it still has a number of blockbuster drugs that remain on patent (based on rough projected global sales for 2008): rheumatoid arthritis drug Celebrex ($2.5 billion, on patent until 2014), Xalatan for glaucoma ($1.8 billion, on patent until 2011), Detrol for overactive bladder ($1.2 billion, on patent until 2012), antibacterial drug Zyvox ($1.1 billion, on patent until 2021) and schizophrenia drug Geodon ($1.0 billion, on patent until 2012).  The company still makes money from off-patent drugs, such as erectile dysfunction treatment Viagra ($2 billion, off patent since 2000), but sales generally drop off dramatically as patents expire and exclusivity ends.  See high blood pressure medication Norvasc ($2.2 billion, but down from $4.9 billion in 2006 since it came off patent in 2007), antidepressant Zoloft ($142 million, but down from $3.3 billion in 2005, off patent since 2006) and antibacterial Zithromax ($121 million, but down from $2 billion in its patent expiration year of 2005).

Pfizer acquires a major player in Wyeth, with blockbuster drugs including antidepressant Effexor ($3.9 billion, off patent in 2010), children’s vaccine Prevnar ($2.7 billion , off patent in 2007 in the US, but extended to 2012 in Europe), arthritis drug Enbrel ($3.8 billion, off patent in 2009), antibiotic Zosyn ($1.3 billion, off patent in 2007) and sales from nutritionals, vitamins and Advil totaling $3.0 billion in 2008.

This represents a trend in the pharmaceutical industry towards consolidation, in the hope of producing the next Lipitor, a drug which can net a company tens of billions of dollars each year.  Of course, monopoly profits only come with exclusivity, so I’m guessing patent lawyers and IP litigators will continue to find themselves very busy, even in the down economy.

Cool Queuing Theory Model Suggests Continuations Not the Problem at the PTO

Recently on Bill Heinze’s I/P Updates, I read about a note published by George Mason authors entitled “Improving Patent Examination Efficiency and Quality, An Operations Research Analysis of the USPTO, Using Queuing Theory“. 17 Fed. Cir. B. J. 133, 163 (2007).    The authors were attempting to validate validate the PTO’s concern with excessive continuation filings and proposed new rules regarding continuations.  The model, which incorporated feedback and branching of applications, was based on two assumptions:

1. A priority queue processes higher priority items first, giving them a shorter queuing delay – but at only at the expense of longer queuing delays for lower-priority arrivals. Since the USPTO gives higher priority to RCEs and continuing applications than to regular new applications, large numbers of RCEs and continuing applications should result in disproportionate waiting times for regular new applications.
2. At the very least, a limit on the number of continuing applications would be desirable because continuing applications can produce multiple generations of unlimited numbers of high priority offspring. RCEs and multiple non-final office actions both contribute to the risk of starvation, but pose less potential risk than continuing applications because they do not produce parallel offspring (branching feedback).

Surprisingly, the study found that continuing applications were not the primary cause of congestion of patent applications at the PTO.  According to the model, the excessive number of non-final rejections per application was the main culprit. The non-final rejections in each round of prosecution exceeded by far the number of second and later RCEs and continuation applications.   Thus, limiting or prohibiting successive generations of RCEs and continuation applications as the only change will not be effective in reducing the backlog.  Instead, reducing non-final rejections per application would be more effective because it would relieve the PTO of its primary burden on the system. 

The note went on to say that hiring additional examiners is not an effective solution because the PTO would need to exponentially increase its personnel in order to keep up with its load of patent applications.  Instead, a policy change addressing the source of the problems is required.

 Thus, it looks like it is time for the PTO to revise its rules, which have already been brought into question by Tafas v. Dudas, described in my blog from Nov. 2.  Using research data and mathematical models may prove to be a valuable aid as the PTO leadership considers its next move. 

Do process steps in a product-by-process claim serve as claim limitations?

This is an old question, but deserving an occasional update:

The Manual of Patent Examining Procedure (“MPEP”) defines a product-by-process claim as “a product claim that defines the claimed product in terms of the process by which it is made . . .” Manual of Patent Examining Procedure (MPEP) § 2173.05(p).  Product-by-process claims were originally limited to circumstances where the product could not be described or distinguished from prior art solely in structural terms, but required reference to the process by which the product was made.  This principle was called “Painter’s Necessity Rule.”  In re Painter.  Today, however, the patent office no longer applies “Painter’s necessity rule” and issues product-by-process claims even where the applicant is able to describe and distinguish the invention solely in terms of composition or structure.  In re Hughes (permitted product-by-process claims even though applicant could describe the invention in structural terms); see also MPEP § 2173.05(p). 

Some district courts have approached analysis of product-by-process claims in infringement suits by explicitly following either Scripps Clinic & Research Foundation  v. Genentech, Inc., or Atlantic Thermoplastics Co. v. Faytex Corp..  A Scripps interpretation fails to limit a product-by-process claim by the process steps, and may render a competitor’s very similar product made by a different process an infringing product.  An Atlantic Thermoplastics interpretation would limit a patented product to the process described in the claim, thus distinguishing a competing noninfringing product from the patented product.  Judge Newman in her dissents states that it is most useful to focus on the facts of the case to determine whether the process limits the process in a product-by-process claim.  See Atlantic Thermoplastics Co., Inc., (Newman, dissenting).  The claims in Scripps were directed to a product that was new and unobvious, and because of its structural complexity, required definition in process terms.  See Scripps Clinic & Research Found.. To Judge Newman, the Scripps claims constituted “true product-by-process claims.”  Atlantic Thermoplastics Co., Inc..  

In contrast, claims similar to those in Atlantic Thermoplastics tend to be directed to relatively simple products that are capable of compositional or structural definition.  See Atlantic Thermoplastics Co., Inc.. Judge Newman refers to this class of claims where the product is old or readily described in structural terms but the process is new as “product of the process claims”.  Atlantic Thermoplastics Co., Inc..  Such products are not properly claimed using the product-by-process format, and thus should be infringed only when an identical or equivalent process is used.  Essentially, Judge Newman’s arguments bring Painter’s necessity rule back into the twenty-first century to apply it to the infringement context.                         

An alternative strategy that can be used independently to avoid choosing between Scripps and Atlantic Thermoplastics, or to support choice of precedent described supra, is to follow the quintessential case on modern claim construction, Phillips v. AWH Corp., and apply traditional construction techniques to product-by-process claims.  Traditional construction techniques involve referring to the specification and prosecution history to interpret patent claims.  For example, the court in Cryovac Inc. v. Pechiney Plastic Packaging, Inc., noted the dissonance in Federal Circuit precedent regarding infringement of product-by-process claims, and instead applied Phillips to find that “[T]he determination of whether the process is properly viewed as a limitation is a context-specific inquiry . . . whether [the claim] contains a product-by-process limitation or not, must be construed in light of the specification.”  The Court found that packing material made from layers oriented coextruded film was limited to the product as accomplished by a racking or blowing process, because the blowing process was described in the specification – both general description and examples – of the patent.    

In addition to the specification, Phillips directs the court and parties to examine the prosecution history of the patent to determine the definition of product-by-process claim terms.  If an applicant points to or adds process limitations to support novelty or nonobviousness or to distinguish prior art, such limitations “act as a limit on the claim for purposes of infringement analysis.”  Kennametal, Inc. v. Cerametal S.A.R.L..  For example, the patent claims in Kennametal, Inc. v. Cerametal S.A.R.L., which were directed to high pressure water jet nozzles, were originally rejected by the Examiner as anticipated by prior art.  In response to the rejection, the patentee amended the claims referring to “a high pressure injection nozzle member” to add the process language “formed by liquid phase sintering without imparting external pressure”.  The patentee stated that the amended claims were not anticipated by prior art because the prior art failed to teach the sintering process without the pressure.  The court recognized that the patentee “inserted the process language specifically to avoid the prior art,” and emphasized that “‘if claim scope is relinquished during prosecution on grounds of patentability, the doctrine of prosecution history estoppel provides that the relinquished scope can not (sic) be recovered’ in an infringement action.”              

These specification and prosecution history arguments may not only independently demonstrate that process steps limit its product claims, but also provide support for following Atlantic Thermoplastics and not Scripps.  For example, if a patent owner is able to analyze the composition and structure of the two competing products in order to bring an infringement suit, then is should not need to resort to the production process to support patentability and infringement.  In that case, the patent owner’s claim may more properly be described as a “product of the process claim” instead of a “true product-by-process claim,” evoking Atlantic Thermoplastics and eliminating Scripps as appropriate precedent to follow.

Patent Local Rules

Local rules for patent cases have become all the rage these days.  It all started with the Northern District of California, followed by the Southern District of California, Eastern District of Texas, and the Northern District of Georgia, amongst others.  Some judges in jurisdictions without local rules have even adopted very similar sets of rules for their own use. 

These rules all have common features.  They require early disclosures of Preliminary Infringement Contentions and Preliminary Invalidity Contentions, together with supporting documents.  They require parties to set out a schedule at the Rule 26 conference, deciding the course of claim construction.  Parties must exchange lists of terms to be construed, preliminary constructions, and then submit a joint claim construction statement.  Construction briefs must be written.  Finally, the claim construction hearing (or “Markman Hearing”) is held, where the judge hears both parties’ constructions and supporting evidence, and finally decides on the construction to be used for the rest of the litigation.

The advantage of these patent local rules is that they set deadlines for the completion of claim construction discovery and the hearing, thus pushing the cases along the court docket.  Parties can focus early on what terms they need to research and construe, and what interpretations will work best for their arguments.  Early disclosure ensures cooperation in these endeavors.  Jurisdictions with these patent local rules are thus popular with plaintiffs who want resolution quickly.

However, patent local rules have been criticized because they may focus the issues too early, before the parties ( and the judge) really know what is at stake.  Rules that do not allow changes in construction after preliminary exchanges may be too rigid as parties realize later in the case that they missed important issues.  Further, many of these procedural events occur before the judge is able to comprehend technology involved.  For this, many have recommended neutral tutorial sessions for the judge, or assistance of Special Masters to help the judge learn the technology or science involved in the case.

The Boston Globe recently ran an article by Sacha Pfeiffer regarding the possibility of drafting patent local rules for D. Mass..  Judge Stearns gave positive comments, and a few other judges have responded positively.  The current President of the BPLA, Lee Bromberg, said that applying patent local rules in D. Mass. would make the jurisdiction more attractive for plaintiffs, lowering costs of litigation for local businesses (particularly some of the smaller biotech companies), and also boost business for local hotels, restaurants, law firms, and support as more people come to litigate.  Just look what making patent litigation more attractive did for E.D. Tex.!  It will be an interesting story to follow as rules are drafted and tried out by the D. Mass. judges.      

The Much Anticipated KSR Decision Is Finally Here

Today SCOTUS finally handed down its decision in KSR v. Teleflex and weighed in on the nonobviousness standard. Writing for a unanimous Court, Justice Kennedy, as expected, rejected the Federal Circuit’s current test for nonobviousness – whether there was a teaching, suggestion, or motivation (TSM) to combine prior art references. The TSM test does provide helpful insights into the nonobvious inquiry, however, that does not demand the application of a narrow and rigid standard. Under the new standard the court must determine whether the is more than a predictable use of the prior art elements according to their established functions. Just a preliminary post. More to come.