Home -> Law Blog Directory -> Patent Law Blogs -> The Patent Prospector

Patent Law

: The Patent Prospector

Product by Process

By Gary Odom and Jordan Kuhn

Abbott Labs got into two patent spats with generic drug makers, in different districts, over 4,935,507, which claims a chemical composition by process, the result being an antibacterial drug. In one, where Abbott sued, Abbott lost on noninfringement in summary judgment. In the other, which was started by Lupin as a declaratory judgment action, Abbott failed to secure a preliminary injunction in counter-claim. Both losses were from claim construction. The CAFC took them both on appeal in a single decision, a seminal ruling on product-by-process claims.

Abbott Laboratories and Astellas Pharma v. Sandoz and Teva and Ranbaxy and Par (2007-1400) and
Lupin v. Abbott Laboratories and Astellas Pharma (2007-1446)

Claim 1 claims crystalline cefdinir, using its chemical name, and defining its unique characteristics with powder X-ray diffraction (PXRD) angle peaks... In contrast, claims 2-5 claim crystalline cefdinir, without any PXRD peak limitations, but with descriptions of processes used to obtain the crystalline cefdinir.

The claim construction warm-up, which in this case involved running a few laps -

Because the claims define the patent right, see Innova/Pure Water, Inc. v. Safari Water Filtration Systems, Inc., 381 F.3d 1111, 1115 (Fed. Cir. 2004), naturally "the claims themselves provide substantial guidance as to the meaning of particular claim terms." Phillips v. AWH Corp., 415 F.3d 1303, 1314 (Fed. Cir. 2005) (en banc). But the claims "must be read in view of the specification, of which they are a part." Markman v. Westview Instruments, Inc., 52 F.3d 967, 979 (Fed. Cir. 1995) (en banc), aff'd, 517 U.S. 370 (1996). A patent's specification provides necessary context for understanding the claims, and "is always highly relevant to the claim construction analysis." Phillips, 415 F.3d at 1315 (quoting Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582 (Fed. Cir. 1996)). While equally true in a general sense, sometimes the specification offers practically incontrovertible directions about claim meaning. For example, inventors may act as their own lexicographers and give a specialized definition of claim terms. See id. at 1316. Likewise, inventors and applicants may intentionally disclaim, or disavow, subject matter that would otherwise fall within the scope of the claim. See id.

When consulting the specification to clarify the meaning of claim terms, courts must take care not to import limitations into the claims from the specification. This court has recognized the "fine line between" the encouraged and the prohibited use of the specification. Comark Commc'ns, Inc. v. Harris Corp., 156 F.3d 1182, 1186 (Fed. Cir. 1998). When the specification describes a single embodiment to enable the invention, this court will not limit broader claim language to that single application "unless the patentee has demonstrated a clear intention to limit the claim scope using 'words or expressions of manifest exclusion or restriction.'" Liebel-Flarsheim Co. v. Medrad, Inc., 358 F.3d 898, 906 (Fed. Cir. 2004) (quoting Teleflex, Inc. v. Ficosa N. Am. Corp., 299 F.3d 1313, 1327 (Fed. Cir. 2002)). By the same token, the claims cannot "enlarge what is patented beyond what the inventor has described as the invention." Biogen, Inc. v. Berlex Labs., Inc., 318 F.3d 1132, 1140 (Fed. Cir. 2003) (quoting Netword, LLC v. Central Corp., 242 F.3d 1347, 1352 (Fed. Cir. 2001)). Thus this court may reach a narrower construction, limited to the embodiment(s) disclosed in the specification, when the claims themselves, the specification, or the prosecution history clearly indicate that the invention encompasses no more than that confined structure or method. See Liebel-Flarsheim, 358 F.3d at 908.

Along with the specification, the prosecution history is "intrinsic evidence" of the meaning of the claims, because it "provides evidence of how the [United States Patent & Trademark Office (PTO)] and the inventor understood the patent." Phillips, 415 F.3d at 1317. Although often producing ambiguities occasioned by ongoing negotiations between the inventor and the PTO, "the prosecution history can often inform the meaning of the claim language by demonstrating how the inventor understood the invention and whether the inventor limited the invention in the course of prosecution, making the claim scope narrower than it would otherwise be." Id. "[C]lear and unmistakable" statements during prosecution may also disavow claim scope. Computer Docking Station Corp. v. Dell, Inc., 519 F.3d 1366, 1374 (Fed. Cir. 2008) (quoting Purdue Pharma L.P. v. Endo Pharms., Inc., 438 F.3d 1123, 1136 (Fed. Cir. 2006)). Again owing in part to the inherent ambiguities of prosecution history, the doctrine of prosecution disclaimer only applies to unambiguous disavowals. See id. at 1375.

The word "crystalline" was only defined in the specification to one form, Crystal A.

The '507 patent claims priority to Japanese Patent Application No. 62-206199 (the JP '199 application), which claimed two crystalline forms of cefdinir, "Crystal A" and "Crystal B." The JP '199 application claimed Crystals A and B very specifically, defining Crystal A by three infrared (IR)-absorption wavelengths and sixteen PXRD angles and intensities. In contrast, Crystal B featured five IR-absorption wavelengths and twenty-one PXRD angles/intensities.

Despite using the JP '199 application for priority, the '507 patent's specification differs significantly. Specifically, Abbott (actually Fujisawa Pharmaceutical Co., Ltd., Astellas' predecessor in interest) jettisoned the Crystal B disclosure found in the JP '199 application and crafted broader claims in its prosecution of the '507 patent. Because the JP '199 applications defines Crystal A and Crystal B physiochemically rather than structurally, the forms actually represent subgenuses of crystalline cefdinir. Thus Crystals A and B comprise crystalline forms of varying structures, which in the context of this case means varying levels of hydration.

The JP '199 application establishes unequivocally that Abbott knew and could describe both Crystal A and Crystal B. Abbott could have retained the disclosure of Crystal B to support the broader claims of the '507 patent, but instead disclosed and claimed A alone.

Then, spouting on "proper interpretation of product-by process claims" the whole CAFC piled in, for this section only, en banc (Judge Schall excepted). Or, as Prof. Crouch put it: "en banc sua sponte sub secretum," because the court received no briefing, nor argued the issue.

The majority rooted its ruling in Atlantic Thermoplastics.

This court in Atlantic Thermoplastics construed product-by-process claims as limited by the process.

This rule finds extensive support in Supreme Court opinions that have addressed the proper reading of product-by-process claims. See Smith v. Goodyear Dental Vulcanite Co., 93 U.S. 486, 493 (1877) ("The process detailed is thereby made as much a part of the invention as are the materials of which the product is composed."); Goodyear Dental Vulcanite Co. v. Davis, 102 U.S. 222, 224 (1880) ("[T]o constitute infringement of the patent, both the material of which the dental plate is made . . . and the process of constructing the plate . . . must be employed."); Merrill v. Yeomans, 94 U.S. 568 (1877); Cochrane v. Badische Anilin & Soda Fabrik, 111 U.S. 293 (1884) (BASF); The Wood-Paper Patent, 90 U.S. 566, 596 (1874); Plummer v. Sargent, 120 U.S. 442 (1887); Gen. Elec. Co. v. Wabash Appliance Corp., 304 U.S. 364 (1938); see also Atl. Thermoplastics, 970 F.2d at 839-42 (discussing each of these cases). In these cases, the Supreme Court consistently noted that process terms that define the product in a product-by-process claim serve as enforceable limitations. In addition, the binding case law of this court's predecessor courts, the United States Court of Customs and Patent Appeals (see In re Hughes, 496 F.2d 1216, 1219 (CCPA 1974) (acknowledging that "true product claims" are "broader" in scope than product-by-process claims)), and the United States Court of Claims (see Tri-Wall Containers v. United States, 408 F.2d 748, 751 (Ct. Cl. 1969)), followed the same rule.

This court's sister circuits also followed the general rule that the defining process terms limit product-by-process claims. See, e.g., Hide-Ite Leather v. Fiber Prods., 226 F. 34, 36 (1st Cir. 1915) ("It is also a well-recognized rule that, although a product has definite characteristics by which it may be identified apart from the process, still, if in a claim for the product it is not so described, but is set forth in the terms of the process, nothing can be held to infringe the claim which is not made by the process."); Paeco, Inc. v. Applied Moldings, Inc., 562 F.2d 870, 876 (3d Cir. 1977) ("A patent granted on a product claim describing one process grants no monopoly as to identical products manufactured by a different process."). Indeed, this court itself had articulated that rule: "For this reason, even though product-by-process claims are limited by and defined by the process, determination of patentability is based on the product itself." In re Thorpe, 777 F.2d 695, 697 (Fed. Cir. 1985) (emphasis added).

The Supreme Court has long emphasized the limiting requirement of process steps in product-by-process claims. In BASF, the Court considered a patent relating to artificial alizarine...

After BASF, the Supreme Court continued to emphasize the importance of process steps in evaluating the infringement of product-by-process claims. See, e.g., Plummer, 120 U.S. at 448 ("[W]hatever likeness that may appear between the product of the process described in the patent and the article made by the defendants, their identity is not established unless it is shown that they are made by the same process."); Gen. Elec. Co., 304 U.S. at 373 ("[A] patentee who does not distinguish his product from what is old except by reference, express or constructive, to the process by which he produced it, cannot secure a monopoly on the product by whatever means produced." (footnote omitted)).

Thus, based on Supreme Court precedent and the treatment of product-by-process claims throughout the years by the PTO and other binding court decisions, this court now restates that "process terms in product-by-process claims serve as limitations in determining infringement." Atl. Thermoplastics, 970 F.2d at 846-47. As noted earlier, this holding follows this court's clear statement in In re Thorpe that "product by process claims are limited by and defined by the process." 777 F.2d at 697.

More recently, the Supreme Court has reiterated the broad principle that "[e]ach element contained in a patent claim is deemed material to defining the scope of the patented invention." Warner-Jenkinson, 520 U.S. at 19. Although Warner-Jenkinson specifically addressed the doctrine of equivalents, this rule applies to claim construction overall. As applied to product-by-process claims, Warner-Jenkinson thus reinforces the basic rule that the process terms limit product-by-process claims. To the extent that Scripps Clinic is inconsistent with this rule, this court hereby expressly overrules Scripps Clinic.

The court reaffirmed the legitimacy of product-by-process claims.

Thus this court does not question at all whether product-by-process claims are legitimate as a matter of form. The legitimacy of this claim form was indeed a relevant issue in the nineteenth century when Ex parte Painter, 1891 C.D. 200, 200-01 (Comm'r Pat. 1891), and some later cases were before the Commissioner of Patents. However, this court need not address that settled issue. The issue here is only whether such a claim is infringed by products made by processes other than the one claimed. This court holds that it is not.

Product-by-process claims, especially for those rare situations when products were difficult or impossible to describe, historically presented a concern that the Patent Office might deny all product protection to such claims. See In re Butler, 17 C.C.P.A. 810, 813 (CCPA 1930) ("Process claims are valuable, and appellant thinks he is entitled to them; but it is submitted that he should not be limited to control of the process when the article which that process produces is new and useful."). In the modern context, however, if an inventor invents a product whose structure is either not fully known or too complex to analyze (the subject of this case - a product defined by sophisticated PXRD technology - suggests that these concerns may no longer in reality exist), this court clarifies that the inventor is absolutely free to use process steps to define this product. The patent will issue subject to the ordinary requirements of patentability. The inventor will not be denied protection. Because the inventor chose to claim the product in terms of its process, however, that definition also governs the enforcement of the bounds of the patent right. This court cannot simply ignore as verbiage the only definition supplied by the inventor.

A rare look behind-the-curtain into rationale. This to buck up the ruling in the face of withering dissent.

This court's rule regarding the proper treatment of product-by-process claims in infringement litigation carries its own simple logic. Assume a hypothetical chemical compound defined by process terms. The inventor declines to state any structures or characteristics of this compound. The inventor of this compound obtains a product-by-process claim: "Compound X, obtained by process Y." Enforcing this claim without reference to its defining terms would mean that an alleged infringer who produces compound X by process Z is still liable for infringement. But how would the courts ascertain that the alleged infringer's compound is really the same as the patented compound? After all, the patent holder has just informed the public and claimed the new product solely in terms of a single process. Furthermore, what analytical tools can confirm that the alleged infringer's compound is in fact infringing, other than a comparison of the claimed and accused infringing processes? If the basis of infringement is not the similarity of process, it can only be similarity of structure or characteristics, which the inventor has not disclosed. Why also would the courts deny others the right to freely practice process Z that may produce a better product in a better way?

In sum, it is both unnecessary and logically unsound to create a rule that the process limitations of a product-by-process claim should not be enforced in some exceptional instance when the structure of the claimed product is unknown and the product can be defined only by reference to a process by which it can be made. Such a rule would expand the protection of the patent beyond the subject matter that the inventor has "particularly point[ed] out and distinctly claim[ed]" as his invention, 35 U.S.C. § 112 ¶ 6.

Claims 2 and 5 claimed product-by-process, where the product was "obtainable by" a process.

As per BASF, Abbott's claim cannot capture a product obtained by or obtainable by processes other than those explicitly recited in the claims.

In sum, a patentee's use of the word "obtainable" rather than "obtained by" cannot give it a free pass to escape the ambit of the product-by-process claiming doctrine. Claims that include such ambiguous language should be viewed extremely narrowly. If this court does not require, as a precondition for infringement, that an accused infringer actually use a recited process, simply because of the patentee's choice of the probabilistic suffix "able," the very recitation of that process becomes redundant. This would widen the scope of the patentee's claims beyond that which is actually invented--a windfall to the inventor at the expense of future innovation and proper notice to the public of the scope of the claimed invention.

One of the cases involved infringement analysis under the doctrine of equivalents.

Infringement analysis under the doctrine of equivalents proceeds element-by-element; a generalized showing of equivalency between the claim as a whole and the allegedly infringing product or process is not sufficient to show infringement. See Warner-Jenkinson, 520 U.S. at 29 ("the doctrine of equivalents must be applied to individual elements of the claim, not to the invention as a whole"). The primary test for equivalency is the "function-way-result" or "triple identity" test, whereby the patentee may show an equivalent when the accused product or process performs substantially the same function, in substantially the same way, to achieve substantially the same result, as disclosed in the claim. Graver Tank & Mfg. Co. v. Linde Air Prods. Co., 339 U.S. 605, 608 (1950). But, because "[d]ifferent linguistic frameworks may be more suitable to different cases," Warner-Jenkinson, 520 U.S. at 40, the function-way-result test is not the only test for equivalency. Equivalency may also be proven where the differences between the invention as claimed and the accused product or process are insubstantial. Hilton Davis Chem. Co. v. Warner-Jenkinson Co., 62 F.3d 1512, 1517-18 (Fed. Cir. 1995) (en banc), rev'd on other grounds, 520 U.S. 17 (1997). In no case, however, may the doctrine of equivalents ignore the individual claim elements. See Warner-Jenkinson, 520 U.S. at 40 (requiring "a special vigilance against allowing the concept of equivalence to eliminate completely any such [individual] elements").

Because the claims were limited to Crystal A, "the doctrine of equivalents cannot capture crystals that are not themselves equivalent to Crystal A."

Note that the doctrine of equivalents and bioequivalency are two different things.

While bioequivalency may be relevant to the function prong of the function-way-result test, bioequivalency and equivalent infringement are different inquiries. Bioequivalency is a regulatory and medical concern aimed at establishing that two compounds are effectively the same for pharmaceutical purposes. In contrast, equivalency for purposes of patent infringement requires an element-by-element comparison of the patent claim and the accused product, requiring not only equivalent function but also equivalent way and result. Different attributes of a given product may thus be relevant to bioequivalency but not equivalent infringement, and vice versa. As the Northern District of Illinois observed in the Sandoz case, "[i]f bioequivalency meant per se infringement, no alternative to a patented medicine could ever be offered to the public during the life of a patent." Sandoz PI Order, 486 F. Supp. 2d at 776. Thus, while potentially relevant, the bioequivalency of an accused product with a product produced from the patent at issue is not sufficient to establish infringement by equivalents.

Abbott out of luck affirmed.

Judge Newman, joined by Judges Mayer and Lourie, dissented from the en banc product-by-process ruling. For 39 pages. Her eloquence ample, and her stern logic formidable, as always.

My dissent is directed as much to the court's procedure, as to the substance of the court's decision.

The court has given no notice of this impending en banc action, contrary to the Federal Rules of Appellate Procedure and contrary to the Federal Circuit's own operating procedures. The en banc court has received no briefing and held no argument, although the Federal Rules so require. The communities of inventors, innovators, and the public who may be affected by this change of law have had no opportunity to be heard. The court has received no information concerning the effect on patents that were granted based on this long-established practice, no advice on what kinds of inventions may now lie fallow because they are unprotected. Nor does the court explain its suspension of the standards of judicial process.

Judge Newman argued that the huge flaw in the majority decision regarding product-by-process is a bifurcation of construction for validity versus infringement.

Defying precedent, the en banc court adopts for all situations "the basic rule that the process terms limit product-by-process claims," maj. op. at 17, whether the product is novel or known, and whether or not the new product could not have been fully described by its structure alone. The court eliminates the long-accepted expedient for new products whose structure is not fully known. While the Scripps decision is the only decision that is mentioned as "expressly overruled," maj. op. at 17, Scripps is only one of many cases now discarded.

The en banc majority's response to the dissenters is to state that "the inventor is absolutely free to use process steps to define this product" if its "structure is either not fully known or too complex to analyze," maj. op. at 19, but to eliminate the premise that the inventor thereby obtains a product claim, not a process claim. According to the majority, a patentee can continue to obtain product claims using process descriptors, but such product claims are treated as process claims for infringement. The applicant would still have to demonstrate patentability of the new product as a product (independent of the process), while enforcement of the patent against an identical product would be limited to the infringer's use of the process steps used as a descriptor. For the first time, claims are construed differently for validity and for infringement.

It has been an inviolate rule that patent claims are construed the same way for validity and for infringement. See, e.g., Amgen Inc. v. Hoechst Marion Roussel, Inc., 324 F.3d 1313, 1330 (Fed. Cir. 2003) ("It is axiomatic that claims are construed the same way for both invalidity and infringement."); Amazon.com, Inc. v. Barnesandnoble.com, Inc., 239 F.3d 1343, 1351 (Fed. Cir. 2001) ("Because the claims of a patent measure the invention at issue, the claims must be interpreted and given the same meaning for purposes of both validity and infringement analyses."); C.R. Bard, Inc. v. M3 Systems, Inc., 157 F.3d 1340, 1363 (Fed. Cir. 1998) ("Claims must be interpreted the same way for determining infringement as was done to sustain their validity."); Southwall Technologies, Inc. v. Cardinal IG Co., 54 F.3d 1570, 1576 (Fed. Cir. 1995) ("Claims may not be construed one way in order to obtain their allowance and in a different way against accused infringers."); Beachcombers, International, Inc. v. WildeWood Creative Products, Inc., 31 F.3d 1154, 1163 (Fed. Cir. 1994) ("We have already interpreted the claims for purposes of assessing their validity. The same claim interpretation of course applies to the infringement analysis."); Scripps Clinic & Research Foundation v. Genentech, Inc., 927 F.2d 1565, 1583 (Fed. Cir. 1991) ("claims must be construed the same way for validity and for infringement"); Smithkline Diagnostics, Inc. v. Helena Laboratories Corp., 859 F.2d 878, 882 (Fed. Cir. 1988) ("The claims of the '970 patent measure the invention at issue; thus, the claims must be interpreted and given the same meaning for purposes of both validity and infringement analyses."); see also 5A Chisum on Patents §18.01 (2007) ("A fundamental tenet of patent law is that a claim must be interpreted consistently for purposes of infringement and validity."); id. §18.03[2][h] (collecting cases).

As interpreted for validity, the claims obtained under the expedient of necessity are product claims, and are subject to the requirements of novelty, unobviousness, and all other requirements for new products, independent of how the products can be made. My colleagues hold that these are product claims for validity, but process claims for infringement. Departure from the rule that forbids such deviation requires sound reason, and fuller exploration than the cursory brush-off dispensed by my colleagues.

The majority's change of law simply imposes unfairness as well as legal error on patent-supported advances.

Judge Lourie had his own beef, of historical bent.

I agree that there is substantial Supreme Court precedent that holds that product-by-process claims require use of the recited process for there to be infringement. However, many of those cases applied overly broad language to fact situations involving old products or used vague language that makes it difficult to determine whether the products were old or new. Clearly, however, when a product is old, a product-by-process claim cannot be interpreted as a claim to the product made by any means. The product is old and unpatentable per se. BASF in fact involved an old product. See Cochrane v. Badische Anilin & Soda Fabrik, 111 U.S. 293, 311 (1884) ("It was an old article.").

Judge Lourie would treat product-by-process as essentially product claims.

There is arguably a different situation that should apply to chemical-biological products today than to mechanical products of more than a century ago. When a product is new and the inventor claims it by a process of preparation, I fail to see why the product-by-process claim should not be interpreted as a product claim that can be infringed even when the product is made by means other than that recited in the claim. Supreme Court precedent dealing with old products, while utilizing broad language, does not foreclose that possibility. The Court years ago did not have occasion to consider today's innovations or decide whether a distinction should be made between a new chemical-biological product and an old product made by a new process.

It may be that with today's analytical techniques there is little need for product-by-process claims. After all, claim 1 of the Abbott patent is a claim to a compound, not only by name, but also by certain of its characteristics. A claim to a product defined by its characteristics or properties surely is a proper claim.

However, product-by-process issues still seem to come before us and I would make a distinction between old products and new products in interpreting product-by-process claims.

Judge Lourie's metric would be rent with the foolery that Judge Newman derided, requiring first validity analysis to determine "a distinction between old products and new products in interpreting product-by-process claims." A legal Schrödinger's cat, quite peculiar that Judge Lourie sat on both sides of the same fence.

Full post as published by The Patent Prospector on May 20, 2009 (boomark / email).