Five years ago, in May 1998, the New York Times ran a story on the efforts of Carl Malamud to make the contents of the U.S. patent database freely available. The rest is history, and it is hard to imagine a world without general, ready, and free access to the texts of U.S. patents. Things are different because of the actions of Carl Malamud. Also in May 1998, the Wall Street Journal (WSJ) ran a story, “Whatever Happened to the Buckyball?” Five years later we are still wondering. Even though the WSJ tipped us off that there was a severe gap between promises and reality as to fullerenes, the hype continues. Thus, in April 2003, when we read about this “dream material easing into the limelight” for applications in medicine, battery technology, fuel cells, semiconductors, and medicine, we might want to ask: what’s been going on for the last ten years? Looking back into history is a major theme in patent law. We frequently have to ask: what did you know, when did you know it, and how can we tell that you really possessed it?
Two Important District Court Cases on Appeal
Of the significance of intellectual property protection in our economy, Federal Reserve Board Chairman Alan Greenspan recently observed: "If our objective is to maximize economic growth, are we striking the right balance in our protection of intellectual property rights? Are our protections sufficiently broad to encourage innovation but not so broad as to shut down follow-on innovation?" The appeals in two recent district court decisions will allow the Court of Appeals for the Federal Circuit, to which all appeals of patent matters are directed, to optimize innovation in the pharmaceutical industry.
In University of Rochester v. G. D. Searle, et al., 2003 U.S. Dist. LEXIS 3030, the Federal Circuit will have the opportunity to tackle so-called "reach-through" claims, claims which attempt to reach-through from one discovery to obtain patent protection on a different discovery not yet made. In the area of non-steroidal anti-inflammatory pain relievers (for example, aspirin, RELAFEN), researchers at the University of Rochester had a theory that two enzymes were implicated in inflammatory pain, and that inhibition of one of them (COX-2) but not the other (COX-1) would be beneficial in treating humans. They identified the portion of the genome coding for COX-1 and COX-2, and were able to make these two enzymes. However, they did not make, or suggest how to make, a drug which was capable of selectively inhibiting the COX-2 enzyme. The University of Rochester obtained a patent claim for a method of treating humans with a drug capable of selectively inhibiting the COX-2 enzyme, and they asserted the patent the day it issued. However, in the litigation, this claim was invalidated because of the failure of the patent to describe such a drug or a way to make such a drug. The defendants in that litigation had made a drug capable of selectively inhibiting the COX-2 enzyme, and were selling it to the public (CELEBREX). One suspects that the decision will be affirmed by the Federal Circuit. From a patent law standpoint, 35 U.S.C. section 112 requires that the patent applicant provide a written description of the invention to show what it is that the inventor possesses and that the applicant show that the invention actually works. Patent law expresses the requirements of a deal between the inventor and the public. The public, in return for the grant of a right to exclude others for a limited period, is entitled to receive an invention which can be put into practice without undue experimentation. In the case of the University of Rochester patent, the public did not receive knowledge of how to obtain a drug to selectively inhibit the COX-2 enzyme, and thus could not use the invention as claimed. The public cannot live on promises of what could have been if some third party did significant work. Of Chairman Greenspan's issues, one notes that a determination that reach-through claims were valid would discourage the innovative effort of those who would actually bring a tangible, working product to the public. Economic growth, in the form of patients buying newly developed, viable drugs to foster better health, would be thwarted.
In the case SmithKline Beecham v. Apotex, 2003 U.S. Dist. LEXIS 2902, the district court addressed the issue of de minimis infringement and concluded that a composition of matter claim can be read as if it excludes any [measureable] amount so small as to lack any commercial significance. Although this interpretation is helpful to generic companies seeking to design around proprietary drugs, it is not clear that it will stand up to text in Embrex v. Service Engineering, 216 F.3d 1343, 1352-1353; 55 USPQ2d 1161 (Fed. Cir. 2000)[J. Rader concurring: "no room remains in the law for a de minimis excuse," the patent statute gives no leeway to excuse infringement because the infringer only infringed a little and that questions of degree go to damages, not to liability]; Madey v. Duke University, 307 F.3d 1351 as quoted in Semitool v. Ebara, 2002 U.S. Dist. LEXIS 21939 ["Similarly, our precedent does not immunize any conduct that is in keeping with the alleged infringer's legitimate business, regardless of commercial implications."]. Provided that the patent holder did meet the requirements of the patent law, the patent holder is given the right to exclude a competitor from making, using, selling, or offering to sell his claimed invention, without reference to the "amount" of the claimed invention to be used. The concept that there can be a "de minimis" level of infringement will likely be rejected by the Federal Circuit. At the same time, however, it should be made clear that there is symmetry between the standards applied for infringement and invalidity. If these same minor amounts were found to exist in a product that existed before the patent, then the patent claim should be found invalid. As the Supreme Court said in Peters v. Active Manufacturing Co., 129 U.S. 530, 537 (1889): "[t]hat which infringes, if later, would anticipate if earlier." In terms of Chairman Greenspan's policy concerns, we will allow the competitor to design around, but if the competitor comes so close that if he makes some of the claimed product, when earlier processes made none of the claimed product, then the competitor is deemed an infringer.
The Glaxo decision has many interesting aspects. The claim at issue simply recites crystalline paroxetine hydrochloride hemihydrate. The words “paroxetine hydrochloride hemihydrate” describe chemical composition. One of ordinary skill knows what paroxetine is, what hydrochloride is, and what hemihydrate means. The word “crystalline” establishes a requirement of a degree of translational periodicity of the chemical entities; in practical terms, it means relatively narrow lines in x-ray diffraction. If this claim is indefinite, there are many, many claims which are indefinite. The opinion repeatedly talks about “single crystals,” although it does not mean “single crystals” as a scientist would understand them (a phase consisting of one crystal manifesting narrow diffraction lines corresponding to long coherence lengths through the Scherrer equation, as distinct from a polycrystalline material having crystallites of different orientations). In the context of the opinion, a single crystal means a single particle of the claimed material. There is much discussing of the “disappearing polymorph” phenomenon, which is more in the realm of anecdotal stories than well-researched evidence. If SARS were as contagious as the alleged seed contamination were pervasive, there would be no one left to read this article. The bigger problem in the case, however, is the adoption of a “de minimis” infringement exception in light of Federal Circuit cases to the contrary.
Apex v. Raritan Vacacted
The district court decision, discussed in our article in the January 2003 issue of Intellectual Property Today in the context of non-response to reasons for allowance, was vacated and remanded on April 2. Curiously, the CAFC did not investigate the propriety of drawing an inference from an unresponded to statement of reasons for allowance, but rather found that the reason in this case was a “standard” reason for allowance which merely said the prior art did not teach or suggest the claimed arrangement. One might infer that if the reason for allowance stated that patentability was based on X, and that if the applicant did not challenge this, the applicant might be bound by the failure to object.
In Passing
Judge Newman’s dissent in Hoffman-LaRoche v. Promega is spirited. Text in Allergan v. Alcon suggests we might have an en banc hearing as to the 271(e)(2) Hatch-Waxman issue shared between Allergan and Warner-Lambert v. Apotex, 316 F.3d 1348 (CAFC 2003). The University of Michigan patent related to breast cancer is stirring up some controversy. Dr. Joseph DeRisi of SARS fame had submitted his virus chip for patenting at University of California, San Francisco, and was turned down. Microsoft and Lucent have become embroiled in a patent war. There is a report that a certain game maker filed for a trademark on “shock and awe.”
End Notes:
1. The opinions expressed herein are to foster academic discussion and debate, and should not be imputed to any employer of the author, past, present, or future. The author may be reached at C8AsF5@yahoo.com, ebert@email.com, or 609-514-5951. The author obtained a Ph.D. from Stanford University and a J.D. from the University of Chicago, and he has written articles cited by the U.S. Supreme Court and by a Nobel Laureate. The title of this article is a variant of “(There’s) Always Something There to Remind Me,” written by Burt Bacharach and Hal David, and performed by (among others) Dionne Warwick. The author notes the sad murder of attorney Delores Marie Harrison in Hamilton, New Jersey in April.
2. New York Times, p. D6 (May 4, 1998), cited within L. B. Ebert, Intellectual Property Today, June, 1998.
3. Wall Street Journal (May 4, 1998). On March 15, 2003, the New Scientist ran a story subtitled “Whatever happened to nanotubes?” Adopting a variant of the George Aiken approach [Declare victory and get out], the author (V. Jamieson) asserted that nanotubes were everywhere, by referring to applications known long before the work of Sumio Iijima of NEC (e.g., S. Iijima, Nature, 1991, 354, 56) and thus relying on the uncertainty of the scope of carbon nanotube. As non-limiting examples of earlier work, carbon filaments grown by means of metal catalysts were discussed in the 1970's by R. T. K. Baker (who used controlled atmosphere electron microscopy (CAEM) to videotape the growth in real time) and tubular carbon forms were known in Boudouard carbon in the 19th century. The present author co-authored a paper: “Electron Diffraction of Carbon Nanotubes,” J. Phys. Chem., 1995, 99, 9639, discussed the carbon/lithium system used as an anode material in battery applications in laptops and camcorders, Carbon, 1996, 34, 671, reviewed early patents in the fullerene area in Carbon, 1995, 33, 1010, and reviewed various books in the fullerene area in Carbon, 1995, 33, 1007 and Carbon, 1997, 35, 437. Separately, the use of carbon nanotubes as cathodes in displays has been advocated by (among others) SI Diamond/Applied Nanotech who held a webcast on April 10, 2003 to discuss the re-issue of U.S. 5,773,921.
4. The Japan Times, “Carbon molecule inspires hope in medicine, energy sectors,” (April 2, 2003). Part of the answer to the approach may be found in “Japan: a tiny leap forward; nanotech may revive its industry,” Business Week, April 14, 2003. Describing work on carbon nanohorns and possible applications, the article emphasizes Japan’s efforts to become a nanopowerhouse.
5. “Greenspan: intellectual property rights problematic,” Reuters, April 5, 2003, summarizing remarks delivered by satellite to economic conference at Sea Island, Georgia. Greenspan also once purportedly said: “[t]he entire structure of antitrust statutes in this country is a jumble of irrationality and ignorance,” as quoted by R. A. Levy, Microsoft and the Browser Wars, 31 Conn. L. Rev. 1321 (1999).
6. For a more thorough discussion of the concept of reach-through claims, S. G. Kunin, et al., 51 Amer. Univ. L. Rev. 609 (2002), available www.wcl.american.edu/journal/lawrev/51/kunin.pdf. Contrary to a letter in Barron’s (April 7, 2003), the issue is not whether the drug companies who formulated actual drugs made use of Rochester’s research, but whether Rochester’s research was commensurate with what was claimed. See also discussion in Geoff Dyer, “Patent System Stifling Research,” (April 15) quoting John Enderby, vice-president of the Royal Society: “If patents are granted that are too broad in scope, they block other researchers from carrying out related work and so hold up development of medicines. This is tremendously bad for science, but the ultimate losers are the patients.”
7. Although not presented by the facts of the Rochester case, there is another issue to contemplate. At one point, there was a “one gene, one protein” hypothesis, wherein it was believed a given sequence on the gene coded for only one protein. Thus, a patent applicant, with the gene sequence in hand might attempt to predict a protein (so-called in silico biology), and thereby both satisfy the utility requirement and reach through to the protein, even though the utility and the protein might be guesses. It is now known that many genes produce multiple proteins via alternative splicing of mRNA transcripts, showing that there are many more proteins in human cells than there are genes in the human genome. Recent work by Dr. Christian Rohlff on proteins from Alzheimer’s patients showed the relationship between gene, transcript, and protein was not 1:1:1 as previously theorized but more like 1:5:10. That is, each gene has five transcripts, which can translate into 10 different protein isoforms in the human tissue. [Separately, note the five different versions of tropomyosin, a structural protein; different cells splice the 11 exons of the one gene differently to produce different protein forms.] Furthermore, a protein may have more than one function. Accordingly, proteomics is a much more complicated area than genomics. As a separate, but related, problem, 2-D gel electrophoresis, the standard for protein profiling [the term proteome was first proposed at the 1994 Siena 2-D Electrophoresis Meeting], typically identifies proteins that are not membrane based. However, approximately 95% of existing drugs today act against membrane-based proteins. Thus, there is a disconnect between current research on primarily non-membrane based proteins and previous work on drugs affecting membrane based proteins.
8. L. B. Ebert, “Increasingly Aggressive Efforts at Patent Enforcement,” Intellectual Property Today, p. 22, June, 2000, incorporated by reference.
9. The defendants in the case argued a failure to comply with the written description requirement of 35 USC 112 P1. The court noted that the compound required to practice the claimed method was described only in terms of its function, and wherein the only means for finding the compound was a trial and error process. The court did refer to Univ. of California v. Lilly, 119 F.3d 1559 (CAFC 1997) and to Enzo Biochem v. Gen-Probe, 296 F.3d 1316 (CAFC 2002), and observed that a mere wish or plan does not satisfy the written description requirement. The Rochester court stated: “Scientific discoveries, and theories based on these discoveries, frequently lay the groundwork for later inventions, but that does not make the discoverer an inventor as well.” The court separately took up the issue of enablement, and determined that a person of ordinary skill would have to engage in undue experimentation with no assurance of success. Of claiming by function, one might also contemplate General Electric v. Wabash, 304 U.S. 364 (1938), United Carbon v. Binney Smith, 317 U.S. 228 (1942), and In re Fuetterer, 319 F.2d 259 (CCPA 1963).
10. There is still disagreement among judges of the Federal Circuit as to the proper scope of the written description requirement; compare the separate concurring opinions of Judge Rader and Judge Bryson in Moba B. V. v. Diamond Automation Inc. Judge Rader believes that the written description requirement is limited to priority protection.
11. Note In re Seaborg, 328 F.2d 996 (CCPA 1964). Applicant argued that a small amount of an unknown, unconcentrated isotope would have been undetectable, and the CCPA concurred. However, what result if a new analytical method were subsequently developed which made that small amount detectable? Contemplate text from L. B. Ebert, “Inherent difficulties,” Int. Prop. Today, Nov. 1999: To illustrate one issue in the interpretation of Continental Can, consider the discovery of buckyballs. In 1984, workers at Exxon disclosed the existence of C60, with the formula as determined by mass spectrometric measurement n6 and a way to make said C60. (J. Chem. Phys., 1984, 81, 3322). They did not disclose the correct structure of C60; Professors Smalley and Kroto proposed the structure of C60 to be that of a truncated icosahedron in 1985. This was not known to be correct until after the work of Huffman and Kratschmer in 1990. Now, assuming that the synthesis of Exxon always led to C60 with the icosahedral structure, would a patent claim to C60 in 1990 (with the truncated icosahedral structure) be inherently anticipated by the Exxon work in 1984, even though no one knew the structure in 1984, and in fact could not have proved the structure until 1990? Assuming that one of ordinary skill would recognize the inherent property, albeit at a later date, one might think there would be inherent anticipation. (Of course, a 1990 claim to merely C60 would be explicitly anticipated by the written description and enablement of the 1984 reference.)
12. The opinion states: “Were claim 1 to be interpreted as broadly as SmithKline contends, it would fail for indefiniteness.” The opinion cites to General Electric v. Wabash for the proposition that members of the technical community must be on notice of the scope of the patent so that they don’t infringe it inadvertently. The opinion reads in from the specification an inference that a competitor would not think the claim covered an amount so minute as to be of no therapeutic, manufacturing, or other commercial significance, and indeed undetectable by any means known at the time. The opinion notes “There is no hint of such a possibility in the patent,” although it may be that there are few patents with composition of matter claims to a single phase that explicitly set a lower bound, in the specification or in the claims. Whether the opinion makes practical sense is one thing; whether it is supported by the language of 35 USC 271 and CAFC precedent is another. Separately, there is an issue with the idea that the amount at issue here might indeed have been undetectable by any means known (at the time of filing). Back in 1983, using a Siemens Type F diffractometer, Joseph Scanlon and I were able to detect, via x-ray diffraction, amounts of material at less than 1 weight percent, and we published the work, first in the Preprints of the Petroleum Division of the American Chemical Society, 1983, 1353, and then elsewhere.[see also endnote 7 of L. B. Ebert, Abstracting Festo, Int. Prop. Today, Oct. 2001] We also reported on diffraction from optically isotropic liquids, work which was subsequently cited by workers using much more intense synchrotron beam at the European Synchrotron Radiation Facility in Grenoble, France. Synchrotron beams, and rotating anodes, which allowed detection of smaller amounts of material than conventional x-ray sources were available in the 1970's to patentees and their competitors. In theory, all of these approaches could detect 1 or 2 percent of a crystalline phase.
13. Refer to endnote 3 of L.B. Ebert, “Reverse Doctrine of Equivalents: Infringement Defense as to Drug Polymorphs,” Int. Prop. Today, February 2001.
14. Apex Inc. v. Raritan Computer, 187 F. Supp. 2d 141 (SDNY 2002).
15. L. B. Ebert, “Just the facts, ma’am,” p. 9, Int. Prop. Today, Jan. 2003.
16. As a separate matter, some of the discussion of inducement under 271(b) looks almost like an advisory opinion.
17. Peter Gorner, Profit motive feared in patenting of breast cancer cells; U. of Michigan defends actions, Chicago Tribune, page 1, Sunday, March 16, 2003. Note that Fawwaz Ulaby of the University of Michigan criticized the Gorner article in the Chicago Tribune on April 7 and in the Detroit News on April 8. However, the delay of five years by the University of Michigan suggests that we may need to re-think the timing analysis of Professor Merges for commercial activities by academics. Delays in publicizing publicly funded work can be much longer than anticipated by Merges and can have far greater impact. See L. B. Ebert, Implicitly Zurko, Int. Prop. Today, p. 22, July 1999. The Michigan work also undercuts the assertion of Maebius and Wegner that university researchers “publish their scientific results immediately.” See L.B. Ebert, The Things We Lean On Are Things That Don’t Last,” Int. Prop. Today, p. 26, Feb. 2003. Separately, see Tom Abate, Critic of biotech corn fears UC won’t give him tenure, San Francisco Chronicle, March 23, 2003.
18. Paul Elias, Gene chip helps find cause of mystery illness [SARS], AP, p. B1 Trenton Times, April 7, 2003.
19. Don Clark, Microsoft makes latest move in battle over Lucent patents, Wall Street Journal, B6, April 11, 2003. Separately, note page 1 article on Lilly’s antipyschotic Zyprexa.
20. Seattle Times, April 14, 2003. Sony filed an application for use as a title in a videogame. The PTO reported more than a dozen applications, for uses including fireworks, lingerie, shampoo, and baby toys.