The last thing a business needs is to invest heavily in obtaining a patent on a product, only to see the patent invalidated when they attempt to enforce it against an infringer. Because of a statutory provision commonly known as the “on sale bar”, this can happen if a business is not extremely careful about engaging in commercial transactions involving their new product prior to filing a patent application. And the task of avoiding the on sale bar has become even more problematic since the advent of the Leahy-Smith America Invents Act (AIA), which added language to the provision that is yet to be definitively interpreted by the courts.

Let’s first consider the pre-AIA version of the on sale bar, 35 USC §102(b), which applies to any US patent application filed before March 16, 2013. It bars patenting any invention that was “in public use or on sale in this country, more than one year prior to the date of application for patent in the United States.” As interpreted by the US Supreme Court, the pre-AIA on sale bar involves a two part test: (1) the invention must be the subject of a commercial offer for sale, and (2) the invention must be ready for patenting. Under Federal Circuit Court rulings, moreover, a commercial offer for sale sufficient to create an on sale bar can include sales between a supplier and a manufacturer. So if a US business has designed a new product and places a purchase order with its supplier for an initial production run, the one-year clock begins to run as of the date the purchase order is confirmed by the supplier. And this remains true even for a foreign supplier, as long as the product is delivered to the US.

With respect to the second prong of the pre-AIA on sale bar test, Federal Circuit decisions hold that an invention can be considered “ready for patenting” if it is either (a) reduced to practice, or (b) depicted in drawings and/or specifications sufficient to enable a skilled person to practice the invention. An invention is “reduced to practice” when a working prototype is made. Therefore, an initial production run of a product, no matter how small, will satisfy the test of being “ready of patenting”, unless the products are used only for experimental purposes.

For patent applications filed on or after March 16, 2013, the post-AIA version of the on sale bar, 35 USC §102(b) applies. The new provision bars patenting any invention which, more than one year prior to the effective filing date, was “in public use, on sale, or otherwise available to the public.” I have added italics to indicate the newly added phrase, which, as of this writing, the courts have yet to interpret. The problem is that the new language can either be read as expanding the class of events that will give rise to an on sale bar or constricting it. Although some commentators have suggested that the post-AIA on sale bar applies only to sales to the public, as opposed to sales from a supplier to a manufacturer, that reading cannot be assumed at this point.

Until the courts speak definitively on the scope of the post-AIA on sale bar, the best practice for businesses to follow is file a patent application, even if only a provisional one, before engaging in any type of commercial transaction that could be interpreted as a sale.

Here’s a real nightmare scenario, but one which actually happened to one my clients recently. They’d been manufacturing and selling a product line for almost a decade, but were shocked to learn that one of their competitors had pending patent applications which, if approved, would give them exclusive rights to the product design. “What can we do to prevent these patents from issuing?!” was the client’s somewhat panicked email inquiry to me.

While the patent statutes and rules provide for pre-issuance opposition to the grant of a patent, the protest must either be initiated before the patent application is published or with the express consent of the applicant. This presents something of a quandary, since a patent application is not publically accessible until it’s published, usually 18 months after filing. On the other hand, waiting until after a competitor’s patent issues to challenge its validity is not a good option for many businesses, since post-issuance review often involves lengthy and costly trial litigation.

I advised my client to consider the more cost effective approach of preparing a pre-issuance Third Party Submission under 37 CFR §1290. This provision allows a third party, that is, someone other than the patent applicant, to submit documents to the patent examiner reviewing a pending application. Submissions may include any patents, published patent applications, or other printed publications that are potentially relevant to the examination of the application. Accompanying the submission must be an explanation of how each document relates to the claims made in the patent application. Since the window for making a Third Party Submission can close six months after the application is published, a certain amount of vigilance is required in monitoring the competition’s patent filings.

Printed publications eligible for submission under this procedure include non-patent publications, such as internet web pages. Fortunately for my client, they had a long-standing website that described their product line in some detail. But we also needed to submit evidence that this website was published before the filing date of the competitor’s patent application. For that purpose, we referred to a web archive to retrieve the client’s archived web page predating the application filing. By the way, the most comprehensive web archive goes by the name of the “Wayback Machine”, borrowed from the vintage cartoon series “Mr. Peabody’s Improbable History”.

Thomas Jefferson reputedly once observed that the price of liberty is eternal vigilance. Something like eternal vigilance is also vital to any business that doesn’t want to see its profitable product lines shut down by a competitor’s patent activity. The cost of retaining patent counsel to help monitor the competition’s patent filings can be small compared to the potential downside of remaining oblivious.

The following was presented at the Joint Patent Practice Seminar, April 2014. Read part 1 here and part 2 here.

Restriction vs Election of Species

While Circuit Judge Alan Lourie – coincidentally our luncheon speaker today – agreed with the majority’s conclusion in denying Safe Harbor status to the Janzen patent, he remarks that “the majority opinion overcomplicates the analysis of this appeal and improperly commingles restriction practice with election of species practice”. By focusing strictly on the clear lines of demarcation set down by the device-method restriction requirements of the parent and grandparent applications, the concurring opinion avoids the morass of the majority’s “multi-level” consonance approach.

In the parent application that preceded both the Janzen patent and the ’498 sibling patent, the patent examiner, beyond making a restriction requirement, also required an election of species from what were referred to as patentably distinct species A, B, and C, stating “the claims shall be restricted if no generic claim is finally held to be allowable. Currently no claims are generic.” Such a requirement for election of species is not the same as a restriction requirement. It is tentative and its consequences are avoidable by the applicant obtaining the allowability of a generic claim encompassing the various species or by filing separate applications to pursue the nonelected subject matter. In my view, the district court and the majority err in even considering the effect of the requirement for election of species in this case.

Judge Lourie aptly notes that, pursuant to 37 C.F.R. §1.146, a species election is not equivalent to a restriction requirement. An election of species is provisional and subject to the allowance of a linking generic claim, which would result in rejoinder of the non-elected claims.   Consequently, the majority’s “multi-level” consonance analysis lays down a maze of demarcation lines that can shift during the course of the prosecution of the restricted patent. Under such an erratic reckoning, the Harbor is anything but safe.

A Word to the Wise

Practitioners would be well advised not to extend the majority reasoning beyond the facts of this case, as it’s likely to prove intractably cumbersome, particularly in situations in which viable generic claims are in play in the restricted applications. Over time, a series of decisions distinguishing St. Jude will likely evolve the Safe Harbor doctrine closer to the version propounded in Judge Lourie’s concurring opinion.

The following was presented at the Joint Patent Practice Seminar, April 2014. Read part 1 here.

1. Lineage and Consonance

Patent jargon is full of colorful genealogical metaphors; and so we have patent “families”, “parent” and “child” applications, and even “lineage”. Since, in the biological sense, none of us can choose our lineage, we consider it unfair for any of us to suffer prejudice because of it. And a very similar ethical sensibility applies to the Safe Harbor doctrine. Since the patentee cannot control the lineage of a patent resulting from a restriction requirement, the sanctions of double patenting should not apply. In one of its few glimmers of lucidity, the otherwise turbid verbiage of the Safe Harbor provision speaks to applications filed “as a result of” a restriction requirement. Interpreting this pregnant phrase, the seminal court rulings hold that patents issued “as a result of” a restriction requirement can’t be cited as references against each other for double patenting purposes. These decisions have developed a logical framework for the Safe Harbor based on a lineal relationship between the “restricted” application, the “reference” application, and the “challenged” application.

If the first pier of the judicially restructured Harbor is “lineage”, the second pier is something called “consonance”. Not only must the “restricted”, “reference” and “challenged” applications all share a common lineage, but they must all honor the lines of demarcation drawn by the original restriction requirement. While acknowledging that consonance was the “most challenging issue” in the St. Jude case, the trial court nevertheless stumbled over the complex lines of demarcation imposed by both restriction and species election requirements. Let’s see if we can understand why.

2. Multi-Level Consonance?

St. Jude Medical was seeking enforcement against Access Closure of the Janzen patent (7,008,439), which taught a particular device for plugging a vascular puncture, as well as an associated method. The jury at trial had found that Janzen’s claims were not patentably distinct from those of another St. Jude patent (5,725,498), which was a “sibling” of the Janzen patent. On that basis, the jury found that the Janzen claims were potentially invalid for obviousness-type double-patenting over the “sibling” patent, subject to the trial court’s determination of the applicability of the Safe Harbor provision. Janzen and its sibling were both continuations of a common parent (5,830,130), which issued on an application filed as a divisional of a grandparent (5,391,183).

In both the grandparent and parent applications, a restriction was imposed between two groups of claims: Group I, drawn to devices for plugging vascular punctures, and Group II, drawn to methods for plugging vascular punctures. Along with the restriction requirements came a requirement for election between three apparatus species, designated as A, B and C. In the grandparent and parent applications, St. Jude elected Group I (device) and Species B. The Janzen patent was issued with both device (Group I) and method (Group II) claims involving a Species C apparatus.

Access Closure argued that the relevant line of demarcation established in the parent and grandparent applications was between Groups I and II, devices and methods, and that Janzen claims, by straddling that line, were not entitled to Safe Harbor protection. On its side, St. Jude propounded a novel theory of “multi-level” consonance, according to which the requirement to elect species – and even sub-species and sub-sub-species – gives rise to multiple layers of “non-elected subject matter”, all of which is fair game for subsequent divisional and continuation applications. In this reckoning, the lines of demarcation merge into a hopelessly convoluted matrix from which double-patenting-exempt applications can proliferate almost endlessly.

Surprisingly, both the District Court and the Fed Circuit majority bought into this Byzantine approach. Their respective consonance analyses differed, leading to reversal of the District Court’s judgment in favor St. Jude, only because the Fed Circuit interpreted the scope of the sibling patent claims more broadly, as extending across methods (Group II) pertaining to all three apparatus species. The Fed Circuit outcome retains a measure of rationality only because they, somewhat arbitrarily, limit the application of their “multi-level” consonance methodology to the election of species, and not also to sub- and sub-sub-species, as St. Jude urges in its brief. Logical consistency in this misguided dispensation would actually have resulted in as many as 240 potential protected divisional categories¹ – enough to float an armada of applications into the Safe Harbor!

Part 3: Judge Lourie’s Concurrence »

¹ In the parent and grandparent applications, election was required among 3 species, 8 sub-species and 5 sub-sub-species, thus engendering, along with the two restriction groups, 240 possible combinations.

The following was presented at the Joint Patent Practice Seminar, April 2014

1. Competing Doctrines: Double Patenting vs Restriction

According to the religious lore of Catholicism, St. Jude is regarded as the patron of difficult cases. He certainly lives up to his reputation in the case of St. Jude Medical, Inc. et al. v. Access Cloture, Inc., decided by the Federal Circuit on last year’s 9-11 anniversary.

The Fed Circuit had the unenviable task of grappling with the so-called “Safe Harbor” provision of 35 U.S.C. §121. The Safe Harbor provision, the Court observes with tongue-in-cheek, is “not a model of clarity”. In fact, it’s an example of one of those interminable sentences that could only have been penned by someone steeped in the discipline of patent claims drafting.  It reads as follows:

A patent issuing on an application with respect to which a requirement for restriction under this section has been made, or on an application filed as a result of such a requirement, shall not be used as a reference either in the Patent and Trademark Office or in the courts against a divisional application or against the original application or any patent issued on either of them if the divisional application is filed before the issuance of the patent on the other application.

Short of invoking the Saint’s miraculous intervention, a bit of background on the genesis of this provision is definitely in order here if we are to set a course for the Safe Harbor through the fog of its cryptic prose. Enacted as part of the 1952 Patent Act, it was intended to reconcile a conflict between two fundamental principles of patent law: that there shall be only one patent per invention, and that there shall be only one invention per patent. These two principles express themselves, respectively, in the doctrines pertaining to double patenting and claims restriction practice.

2. Safe Harbor as Collision Insurance

Each of these two competing doctrines has its own equity-based policy underpinnings. On one hand, the proscription against double patenting aims at preventing the over-extension of patent protection by filing successive applications based on obvious variants of the same invention. On the other hand, the imposition of restriction requirements ensures that claims for patentably distinct inventions are prosecuted in separate applications. In a perfect world, these two policies would never clash, but in the real world – as all of us are painfully aware – patent examiners and judges are fallible, and their opinions on what are or are not patentably distinct inventions are apt to differ.

By way of example, an examiner requires that the applicant elect between claims 1-10 and claims 11-20, based on his/her non-appealable judgment that they are directed to two different inventions. The applicant dutifully elects claims 1-10 and then files a divisional application based on claims 11-20. Patents issue on both applications, but in subsequent infringement litigation, the defense of invalidity based on double patenting is raised, and the court concludes that the two groups of claims are not patentably distinct. Despite having steered its applications on the only course allowed by the PTO, the patentee suffers a collision that sinks its claims and works a forfeiture of its rights.

To prevent such manifest injustices, Congress saw fit to create the Safe Harbor, in which restriction-driven divisional applications and their progeny would not encounter the perils of double patenting on the open sea. But the compass of the Harbor needs to be carefully circumscribed, lest it become a channel for by-passing the double patenting proscription altogether. In the St. Jude appeal, the Fed Circuit had the opportunity to set out some marker buoys to better define the limits of the Harbor. Let’s consider how they went about doing that.

Part 2: Drawing Lines of Demarcation »