Courts Cast Clouds Over PCR Pricing

The polymerase chain reaction (PCR)--a technique invented by Kary Mullis in 1983, published in 1986,1 and the subject of the Nobel Prize in chemistry in 1993--can well be described as one of the most important technical advances of modern molecular biology. How much researchers have to pay to use the technology, however, is now largely in the hands of US and European courts that are deciding who controls patents on a critical enzyme that, in simplistic terms, amounts to the P in PCR. Basel, Switzerland-based F. Hoffmann-La Roche and Pleasanton, Calif.-based Roche Molecular Systems Inc. (Roche), along with Applied Biosystems in Foster City, Calif., own the patents that cover the use of Taq polymerase in PCR, as well as the technique itself and related instrumentation. But recent legal decisions in the United States and Europe concerning the patentability of one form of this important enzyme may affect the way it is sold, possibly driving down the price of an enzyme that currently generates, according to Roche, approximately $100 million annually in worldwide sales. At issue is the potential for patents held by Roche on recombinant Taq and on PCR technology in general to fall in the wake of decisions affecting the validity of the patent held on the native enzyme (the polymerase purified directly from the thermophilic bacteria, Thermus aquaticus). Legal opinions on the implications of the decisions are mixed. Although other patents in Europe and the United States covering the gene encoding the enzyme have been issued and are currently active,2 the very closeness of the technologies may bring them into question as well. "If the cases are related--for example, by claiming priority to each other--then these original rulings in the United States and Europe will likely taint the recombinant Taq patents and call into question the validity of these patents. If they were filed completely separately, then the validity of the recombinant Taq patent may not be affected. But I'm not sure how they can't be affected. The subject matter is just too close," explains biotech patent attorney Lisa Haile, a partner in the San Diego office of Gray, Cary, Ware, and Freidenrich and co-chair of the firm's Life Sciences group (Haile's firm is not allied with any of the companies in the Taq polymerase dispute). Attorney Brenda Furlow, who represents Madison, Wis.-based Promega Corp., a principal in the ongoing litigation against Roche's PCR patents, is less circumspect: "The evidence relied on to overturn [the] native enzyme claims are relevant to the recombinant Taq patents," she says. The PCR Patent Debate PCR requires a thermostable DNA polymerase, and although alternatives exist, Taq DNA polymerase has been the mainstay of the process since it was commercialized in 1987. Researchers acquire a license to use PCR technology by purchasing a licensed enzyme and using a licensed thermal cycler, or by purchasing the appropriate PCR research rights directly from Applied Biosystems. Thus the cost of the enzyme, if purchased from a licensed supplier, includes a royalty fee.3 Courtesy of the University of CincinnatiRalph Meyer (at right), University of Cincinnati professor of biological sciences, was an expert witness for Promega in the European patent hearings. In December 1999, Judge Vaughn R. Walker of the US District Court, Northern District of California, ruled that all claims of the Taq DNA polymerase patent held by Roche (US patent 4,889,818, commonly referred to as the '818 patent) were unenforceable due to the inventors' "inequitable conduct" in obtaining the patent.2 The decision has been appealed by Roche; the US Patent and Trademark Office (PTO) declined to comment on the case. More recently, on May 30, 2001, the opposition division of the European Patent Office (EPO) found that the claims in Roche's broad thermostable enzymes patent EP-0-258-017 B1 were unpatentable because the technology lacked novelty. A written opinion by the EPO will be issued later this year. According to Clelia Baur, director of government and public relations for Roche, the company will appeal the decision to the EPO Technical Board. The Opposition Division of the EPO upheld another Roche patent on the gene encoding Taq in January 2000, a decision that is being appealed by London-based polymerase vendor Bioline Ltd. To summarize the background of the ongoing dispute, in 1991, Roche acquired the '818 patent from the now-defunct Cetus Corp. of Emeryville, Calif. In 1990, Cetus had granted Promega a license to manufacture and sell native Taq for non-PCR applications. Although Roche offered Promega a license to sell Taq for PCR research, Promega declined the offer and continued to sell non-licensed Taq for PCR. In 1992, Roche sued Promega for breach of contract. Promega countered by challenging the validity of the patent in 1993. In 1995, Roche issued a list of 200 individual scientists who used Promega's Taq in their published research as "infringers" of the '818 patent. In September 1995, Promega and Roche agreed to separate the breach of license claims from the validity claims and decided to try the patent claims first.4 Promega argued that the patent should not have been issued based on prior art; that is, papers published prior to the filing of the Cetus patent application described purification of the enzyme.5,6 Additionally, Promega claimed that Cetus misled the PTO by intentionally omitting evidence that allegedly demonstrated that the enzymes isolated in the prior art publications were in fact full-length, active Taq polymerase. It was the latter argument--the case for inequitable conduct--that eventually led to the revocation of the '818 patent in 1999. In Europe, the EPO issued Roche a broad thermostable enzymes patent in 1997. A year later, Promega, joined by Franklin Lakes, N.J.-based Becton Dickinson; Beverly, Mass.-based New England BioLabs; and Bioline, filed an opposition with the EPO, claiming prior art invalidity. At question in all of these cases was whether the polymerase described in earlier publications was the same one described in the patent held by Cetus and Roche. Promega claimed that efforts to reproduce the results of the original publications successfully yielded full-length Taq polymerase. On the other hand, Roche argued that the prior publications, which reported smaller molecular weights for the purified enzyme, actually purified a natural proteolytic degradation product of full-length Taq that retains polymerase activity (the "Stoffel fragment"). Additionally, Roche claimed that their own efforts to reproduce the original experiments were not successful in producing full-length Taq. Ralph Meyer, a professor of biology at the University of Cincinnati who provided expert testimony for Promega at the European hearings, explains that, in his analysis, Roche's failure to reproduce the experiments was the result of several technical errors. "When I did the analysis of the notebooks, I noted 25 technical errors that did not allow them to continue or reproduce the experiments," he says. Meyer says that most of these errors were minor, but that some interfered with the purification of the sample. For example, at one point the fraction collector stopped, spilling some of the peak sample on the floor, thus preventing an accurate quantification of final purified product. Meyer had been on thesis committees of Alice Chien and David Edgar, students in John Trela's laboratory at the University of Cincinnati who originally worked on the purification of Taq polymerase. "The Trela group [developed the procedure] because we thought that this was going to be an easy enzyme to purify. Anyone with any kind of technical knowledge should be able to purify this enzyme with no problem," Meyer says. The University of Cincinnati publications that described the purification of Taq polymerase may have been central to the EPO's decision; according to Meyer, the EPO viewed Edgar's thesis as the most important prior art document and based their ruling on the conclusion that this thesis disclosed full-length Taq polymerase and that Taq polymerase was thereby in the public domain. According to Melinda Griffith, Roche general counsel, however, "none of the reproductions of prior art protocols conducted by Promega and its experts faithfully followed the details of those protocols." Also, Griffith argues that subsequent work published in 1988 by the Trela group showed that the enzyme was not the same as that patented by Cetus; this work characterized the Chien enzyme by SDS-PAGE and described the enzyme as being made up of two identical subunits, each only 35 kDa in size.7 Nevertheless, the judge in the US case agreed with Promega that Cetus researchers David Gelfand and Susanne Stoffel knew of studies that showed that size exclusion chromatography, the technique used by the Trela group in their original characterization of Taq, could underestimate the true molecular weight of a hydrophobic protein like Taq and that they did not disclose this to the PTO when they filed the patent originally.2 The central question at this point is whether the rulings in the United States and Europe will have an effect on the validity of Roche's patents covering the gene encoding recombinant Taq polymerase and the production of recombinant Taq. Roche has continued to maintain that the patent in question covers only the native form of the polymerase, which comprises only 10 percent of the total Taq market.8 The patent revoked in Europe, however, included claims to other thermophilic enzymes, including recombinant Taq. The '818 patent included two claims to the native Taq enzyme and one to the recombinant version. Patent attorney Stephen Bent, a founder in the Washington, D.C., office of Foley and Lardner (also not affiliated with either party in the dispute), explains that the recombinant Taq issue will be dealt with differently in Europe and the United States, as the question of inequitable conduct is only applicable in the United States. In Europe, the question of whether the recombinant technology was obvious based on the native Taq publications is one of prior art. However, in the United States, "Once you have committed inequitable conduct, the technology recedes into the background, and so there will be an issue of whether the inequitable conduct identified to date taints other applications that are in the same lineage," he says. The Net Result? When all is said and done, will any of these decisions affect the price of Taq? Several licensed suppliers of Taq polymerase contacted by The Scientist declined to comment on whether the decisions will impact pricing; most of those who did comment requested anonymity. A representative from one company that sells recombinant Taq said that her company is selling the product without a license as a result of the decisions: "Based on what we know about the case... Promega won the right to sell [recombinant Taq] without needing a license from Roche... . We believe that we do not need a license either for selling this product." But this is not necessarily the case with all suppliers. Jae-Hyung You, chief research scientist at Bioneer Corp. in Taejon, South Korea, explained in a written statement that the decision will not have any effect on Bioneer's Taq polymerase sales and that Bioneer would like to see the issue resolved soon, as "unlicensed suppliers have been providing Taq DNA polymerase and PCR mixes at surprisingly low prices... [and] licensees with Roche, including Bioneer, are receiving severe blows from price competition." For example, a 1998 LabConsumer price comparison found that nonlicensed suppliers sold native Taq polymerase for nearly half the cost of the licensed enzyme.3 Courtesy of Promega Corp.Brenda Furlow, vice president and general counsel for Promega. According to one PCR product manager representing a licensed supplier of Taq, what happens to the Taq patent may be moot, because most companies pay royalties on the broader PCR patent (which, according to Griffith, expires in 2005); pricing therefore will not be affected unless the PCR patent is overturned. And is this likely to happen? Former Roche attorney Victor Lee thinks it's a stretch: "[Promega has] gone so far as to say in many of their press releases that by virtue of these decisions... the PCR patent would fall as well... [but] there's simply no basis for that." Haile agrees with Lee's assessment. "I know Promega has made comments that this may affect the enforceability of... all the PCR patents, but I think that might be a little extreme." In fact, the broad PCR patents were upheld by a jury in 1991 in a dispute between Cetus and Wilmington, Del.-based E.I. du Pont de Nemours and Co.3 Nevertheless, according to Promega's Furlow, the recent Taq decisions may affect Roche's entire PCR licensing strategy. "I think that Roche's strategies are called into question, since there's no longer any patent covering the native enzyme," she says. "That's a fair statement," says attorney John Harkins, a founder of Harkins Cunningham in Philadelphia (a firm not representing either Roche or Promega), noting that although the Taq polymerase decisions aren't likely to invalidate the PCR process or instrumentation patents, PCR licensing will become more problematic now that the validity of the native Taq patent has been questioned. However, according to Lori Murray, a representative of Applied Biosystems' licensing office, the recent decisions will have no effect on Applied Biosystems' PCR licensing strategy (additional information about PCR licensing can be found on Applied Biosystems' Web site, www.appliedbiosystems.com/legal/pcrlicfaqs.cfm). Promega's Furlow believes that the fallout of this ongoing battle will lead to "broader access to a very important enzyme at less expense," but the consequences of the US and European decisions, and the results of the various appeals, remain to be seen. And although the issues involved in the dispute are largely isolated from questions such as the patentability of sequences derived from the human genome project, the question of disclosure of prior art is central to biotechnology patents. As the products of biotech research become more profitable, there may be more of a push to patent technologies that were previously in the public domain. The Taq polymerase decisions may be a step toward combating this trend. "To a certain extent the courts are really emphasizing the importance of awarding patents to those who deserve the patents, and really affirming ... the importance of being completely up front with the patent office with any knowledge of any prior findings," explains Haile. Aileen Constans can be contacted at aconstans@the-scientist.com. References 1.K. Mullis et al., "Specific enzymatic amplification of DNA in vitro: The polymerase chain reaction," Cold Spring Harbor Symposia on Quantitative Biology, 51: Pt. 1, 263-73, 1986. 2.D. Steinberg, "Patent wars: Judge voids Taq patent," The Scientist, 14[1]:6, Jan. 10, 2000. 3.S. Beck, "Do you have a license? Products licensed for PCR in research applications," The Scientist, 12[12]:21, June 8, 1998. 4.R. Finn, "Ongoing patent dispute may have ramifications for academic researchers," The Scientist, 10[20]:1, Oct. 14, 1996. 5.A. Chien et al., "Deoxyribonucleic acid polymerase from the extreme thermophile Thermus aquaticus," Journal of Bacteriology, 127:1550-7, 1976. 6.A.S. Kaledin et al., "Isolation and properties of DNA polymerase from extreme thermophylic bacteria Thermus aquaticus YT-1," Biokhimiia, 45:644-51, 1980. 7.J. Verhoeven et al., "Further characterization of a DNA polymerase from an external thermophile, Thermus aquaticus," Abstracts of the Annual Meeting of the American Society for Molecular Biology, Abstract K-47, p. 214, 1988. 8.R. Service, "Roche dealt a setback on European Taq patent," Science, 292[5523]:1815, June 8, 2001. Supplemental Tables Available Thermostable Polymerase Mixtures Thermostable Polymerases (non-Taq) Suppliers of Taq DNA Polymerases Click to view the PDF file: List of Thermostable Polymerases Suppliers

Courts Cast Clouds Over PCR Pricing

Interested in reading more?

Become a Member of