Press the print button on your browser.
Click here to return to the previous page.
Report 9 of the Council on Scientific Affairs (I-00)
Full Text
| Patenting of Genes and Their Mutations NOTE: This report represents information on this subject as of December 2000. Full Text Resolution 510, introduced by the American College of Medical Genetics and the College of American Pathologists and adopted at the 2000 American Medical Association (AMA) Annual Meeting, asks: "That the AMA examine the issues surrounding gene patenting and report to the House of Delegates at the 2000 Interim Meeting." This report discusses the current status of gene patenting, changes proposed by the US Patent and Trademark Office (PTO) Revised Utility Examination Guidelines,1 and the current controversy between the biotechnology industry and the biomedical community. Methods: The report reflects extensive discussions with representatives of industry and specialty societies, as well as review of Council on Ethical and Judicial Affairs (CEJA) Report 2, "Patenting the Human Genome" (I-97), and the June 2000 proceedings of the Department of Health and Human Services (DHHS) Secretary's Advisory Committee on Genetic Testing.2 In addition, MEDLINE and BIOETHICSLINE (National Library of Medicine databases) were searched using the terms gene and patent. Introduction Sequencing of the human genome has been equated in importance with the discovery of germ theory. With the White House announcement earlier in 2000 of the rough draft completion of sequencing of the human genome jointly by the publicly funded project headed by the National Human Genome Research Institute director, Francis Collins, MD, PhD, and Celera, a private company headed by Craig Venter, PhD, the issue of gene patents was highlighted. Questions raised include: Whose genes are these? How can something that occurs in nature be patented? It is estimated that between 50,000 and 100,000 genes are distributed within the nearly 3 billion base pairs in the human genome. The human genome is organized into 23 pairs of chromosomes consisting of DNA and other proteins. Genes consist of DNA sequences that are organized into coding regions (exons) and noncoding regions (introns). (A glossary of genetic terms is available at: http://www.genome.gov/glossary.cfm.) When a gene is transcribed into messenger RNA (mRNA), the introns are spliced out. The remaining exons are exported out of the nucleus toward the endoplasmic reticulum where the gene is translated into protein products. Using recombinant technology, including reverse transcriptase, DNA sequences homologous to the mRNA can be produced as complementary DNA (cDNA). These sequences can be used, among other things, to probe cellular DNA. Only about 4% of the human genome consists of coding sequences. The exact nature and function of the noncoding sequence continues to be explored. Patent Law To understand the current controversy about gene patenting, a brief review of patent law is presented. The US patent system dates to the time of the Constitution, which gives Congress the power "to promote the progress of Science and useful Arts, by securing for limited times to authors and inventors the exclusive right to their Writings and Discoveries." Patent law consists of federal statutes and regulations, with interpretation by the PTO and the federal courts. As well summarized in CEJA Report 2, " Patenting the Human Genome" (I-97), "a patent holder does not own an invention, merely the patent." This gives patent holders the right to prevent others from making or using their invention, but does not necessarily give the patent holder the freedom to "practice their invention." In exchange for this right, patents facilitate the sharing of information since full disclosure of the invention is required in awarding of the patent. The converse of a patent is a trade secret, which does not promote scientific progress by those not in possession of the trade secret. The current term of a new patent is 20 years from the date of filing an application. The information contained within an application enters the public domain within 18 months of filing (a recent harmonization of US patent law with international practice). It is the patent holder's responsibility to enforce the patent by litigating any infringement of the patent by others. A patent is awarded for an invention, not a discovery. An invention must fulfill, among other requirements, three fundamental criteria: novelty, non-obviousness, and utility. Generally biotechnology "inventions" fall into two classes: (1) composition of matter related to newly isolated genes or proteins; or (2) methods of treating patients through the use of a particular gene or protein. Novelty and non-obviousness imply that the invention is indeed new, which requires that the chemical structure or in the case of genes, the sequence, be previously unknown. As further detailed in CEJA Report 2 (I-97), the non-obviousness criteria will change over time as technology moves forward. Utility or usefulness has to be specified, not vaguely implied in futuristic terms. The PTO is currently revising the utility criteria, to make it more rigorous.3.4 The utility criteria cannot be met merely on the basis of a genetic sequence serving as a tool for further research on similar gene sequences, which has implications for the patentability of expressed sequence tags (ESTs) (discussed below). There must be a "credible" utility such as in identification of a function of the gene that has an established utility, or clinical testing or clinical treatment of a specific disease. Additionally, the application must describe the invention in such detail that it could be made and used by others skilled in that area. It is this latter requirement of the patent application that promotes the advancement of technology and at the same time protects the inventor. Any commercial use (eg, in the case of genes, the provision for genetic testing) of a potential nucleotide sequence will require a negotiated licensing agreement with the patent holder. However, not all patents are enforced, as either the holder does not care or does not act because of the cost of enforcement. The responsibility of upholding the patent lies with the patent holder, and legal challenges of patent infringement can be quite costly. More than 2,300 gene patents have been issued, and there are more than 10,000 gene patents pending. Currently the US government holds the largest number of gene patents. Issues Related to Gene Patenting During discussion of Resolution 510 at the 2000 AMA Annual Meeting, the American College of Medical Genetics and the College of American Pathologists raised objections to gene patents. The first holds that genes and their mutations are naturally occurring substances; hence, they can only be discovered, and as such are not patentable. The argument that naturally occurring substances are not patentable has been upheld. However, if a naturally occurring substance is isolated or manipulated from nature it may be patentable. Naked nucleic acids do not occur in nature. DNA molecules within a cell are organized as part of a chromosome or within a plasmid such as mitochondria. The gene for erythropoietin was one of the first genes patented. This gene was isolated by working backwards from the protein product. Currently, there are more than a dozen patents on erythropoeitin. Additional patents exist for new, novel uses of erythropoietin beyond the original "composition of matter" patent. The first patent was issued for isolated erythropoietin protein having in vivo biological activity. Subsequent patents include those for a mutant variant of the protein and the isolated DNA sequence encoding the protein. A review of the patents issued for erythropoeitin demonstrates the ability of subsequent patents to cover new utilities of the original protein and DNA sequence. Such patents respect the original patent holder by requiring that for the new invention to be used, a licensing agreement must exist between the patent holders. Patent licenses are, in essence, contracts between the patent holder and other parties who want to use the invention of the patent holder. In such cases, a licensee is usually required to pay the patent holder a percentage of the revenue generated by the use of the invention. A large percentage of the 10,000 or so pending gene patents are for expressed sequence tags (ESTs). An EST is a complementary DNA sequence synthesized from knowing the coding sequence of a gene or partial sequence of a gene. The sequence is homologous to and is produced by reverse transcriptase activity from the mRNA. ESTs can be used to identify genes on chromosomes or genes from a gene library (eg, commercially available cDNA libraries developed from different organs and/or different species). ESTs also can be used to identify multiple members of a class of genes. Gene patents issued on the basis of ESTs and all the downstream products of such genes may not meet the new proposed standards of utility by the PTO (Revised Utility Examination Guidelines, Docket No. 991027289-9289-01). It is noteworthy that the biotechnology industry is in general agreement with the more rigorous utility guidelines. However, future research may define a more clinically relevant utility for ESTs as therapeutic agents based on demonstrated ability to "knock-out" a specific function by virtue of anti-sense function, which would fulfill the more rigorous standard.5 The industry and the scientific community diverge on the use of computational genomics programs to assign utility on the homology of newly isolated gene sequences. Large databases exist of the nucleic acid sequences of DNA and the amino acid sequence of proteins. A novel sequence can be compared to these databases, and matches found with differing levels of homology (matching). A new sequence may have 90% homology to a tyrosine kinase protein; however, this does not prove definitive functional homology. A single nucleotide difference can change or abolish the function of a protein (eg, the alteration of a single nucleotide that is the common mutation causing sickle hemoglobin). Genentech Senior Vice President of Research, Dennis J. Henner, in testimony provided to the US House Judiciary Subcommittee on Courts and Intellectual Property, said that current computational models are not sufficiently accurate to predict protein function based solely on homology analyses.6 However, proof of utility by computational analyses likely will continue to evolve with better predictive functional capacity. Henner noted that it is Genentech's practice to perform protein expression and basic biologic research prior to commitment of significant research and development investment to a newly isolated gene sequence.6 Single Utility: In practice, a description of single utility is sufficient, and it is unusual that a patent is rejected on the basis of inaccurate utility claims. The National Institutes of Health (NIH) in response to the proposed revisions cautioned about the acceptance of "well established" utility without any further elaboration of the specific utility either from the inventor or that the PTO interprets. For the public record there needs to be complete disclosure of what is meant by "well established" utility. The NIH was further concerned about the language of the written description, another of the required elements of the patent application. Jack Spiegel, PhD, Director, Division of Technology and Transfer, responding for the NIH to Q. Todd Dickinson, Commissioner of Patents and Trademarks, stated that the term "comprising" is much too broad and vague, and risks granting patents with inappropriately broad scope.7 The concern is the same as for the perceived low threshold of the utility criteria. The utility asserted in the application need not necessarily be correct. The patent held by Human Genome Sciences, Inc. (HGS) on the CCR-5 protein claimed that it is a mediator of T cells with potential activity in treating rheumatoid arthritis. Subsequently, AIDS researchers found that it is an important receptor for HIV entry into target cells. However, the original patent may nonetheless be dominant despite the fact that others demonstrated a potentially more important utility. Thus, the discoverers of the more important utility must now negotiate with HGS to develop commercial uses of this invention.8 Other issues raised by gene patenting that have been articulated by the biomedical community include the argument that gene sequence-based test services are part of medical practice and should be covered by the Ganske-Frist legislation of 1995 that bars enforcement of any medical procedure patents against any and all infringers. The legislation was passed in recognition that enforcement of patents granted on medical and surgical procedures was having a deleterious effect on patient care. The original language of this legislation did include biotechnology research, and considerable opposition was raised by the biotechnology industry at that time. The final legislation had no impact on biotechnology. Some members of the biomedical community argue that patents may have a chilling effect on training of health care professionals because of the requirement to obtain licensing arrangement and the payment of licensing fees. The irony is that many of the gene patents held by universities have subsequently been licensed to a company that now offers the testing commercially. In some cases, the licensing agreements have been exclusive, with the result of inhibiting the researchers who developed the test to provide clinical testing. In an informal survey of 74 university-based clinical laboratories, 25% found themselves in this exact position. Forty-eight percent have decided not to develop or perform a test for clinical or research purposes because of patent restrictions.9 The PTO has no jurisdiction over patent licensing agreements. In addition, by granting limited or exclusive licensing on some gene-based tests, clinical data regarding the use of the test, including the subsequent specificity and sensitivity, now become proprietary and may not be available for public scrutiny. Despite the option of obtaining a licensing agreement that requires payment or royalty for use, it is possible that the licensee could be "gagged" regarding findings that question the validity and quality of data. Review of pre- and post-market genetic testing data is an issue that the Department of Health and Human Services Secretary's Advisory Committee on Genetic Testing is grappling with in the context of making recommendations on appropriate regulatory oversight for the introduction of genetic tests into clinical practice. The industry argues that patents are needed to attract the necessary capital investment required to develop gene-based testing and therapies. Without the protection of patents, certain companies claim they would not be able to recoup their research and development costs, and future research and drug discovery would suffer and patients would lose. The biotechnology industry's contribution to basic and applied research is no longer trivial. It is one of the country's major industries, employing a highly trained workforce. The impact of this industry was quite evident in the race that occurred between the publicly funded human genome project and the privately funded effort by Celera that culminated in the joint announcement of completion of the human genome sequence in June 2000. Example of Issues Raised by Gene Patenting: Canavan Disease An account of the Canavan disease gene is illustrative of the problems that have arisen from the awarding of a patent for the gene for this neurodegenerative disease. In testimony provided to the Secretary's Advisory Committee on Genetic Testing in June 2000, Judith Tspisis, PhD, recounted how the families of children affected with Canavan disease organized and approached the research community to encourage study of the disease. Through participation of the families, the metabolic basis of the disease was found and eventually the gene was identified. Given the frequency of this autosomal recessively inherited disorder (1/40 in Eastern European Ashkenazic Jews10), and the new availability of genetic testing, the genetics community recommended that couples of Ashkenazic background be tested as part of prenatal screening akin to prenatal screening for Tay-Sachs. Shortly thereafter, the families were informed by Miami Children's Hospital of its intent to enforce its patent infringement rights by seeking an exclusive licensing partner while signing a licensing agreement with university-based clinical laboratories specifying testing caps. Many laboratories stopped offering the test given the adverse nature of the licensing agreements. The ultimate impact is that the test is currently not available to many of those who desire it. The families who participated in the research now must carry the cost of the test or are unable to access the test, raising serious issues of justice and fairness. This case raises the prospect of costs of multiple licensing agreements being passed on to the patient. In multiplex testing (ie, the testing of several diseases at once in a bundle) of common Jewish genetic diseases, given that genetic testing is not uniformly covered by medical insurance, testing may become unaffordable, thus limiting access to medical care. In some cases, additional research findings are generated by use of testing in clinical settings, and may represent an infringement of the patent. If exclusive licensing agreements inhibit further clinical studies on the utility of the test, the result will be lowering of the quality of medical care. If awarding of patents discourages the further refinement of a genetic diagnostic test such that it detects more mutations, eg, increasing the sensitivity of finding rare mutations in common disorders, that would again lower the quality of care. On the other hand, it is also possible that without the protection of patents, tests for relatively rare disorders (orphan diseases) would not be developed or made available because biotechnology companies would not have the financial means to justify research and development.11 As unfortunate as the Canavan story is, the issue does not lie with the issuance of the patent but in the subsequent licensing agreement, which is not the focus of the PTO. Lee Bendekgey, JD, General Counsel of Incyte Genomics, has recommended that gene patenting policies meet several criteria, including that royalty and transaction fees be low to encourage the wide availability of diagnostic tests.12 Conclusion Many of the controversies about current patent law reflect broader issues of health care policy--in particular, the issues of license agreements that inhibit use, too greatly increase the cost, or inhibit research on the patented gene--and are therefore not the purview of the PTO. However, the requirement of the patent application to demonstrate utility; the biomedical community' s concern about the low threshold to meet that requirement; and the potential impact of inhibiting the full exploration of patented genes that might occur later through subsequent clinical application, which is of concern to maintain a high quality of medical care, are issues that must be addressed. Informal survey data that demonstrate the withdrawal of genetic testing by university-based laboratories need further study. RECOMMENDATIONS The following statement, recommended by the Council on Scientific Affairs, was adopted by the AMA House of Delegates as AMA Policy at the 2000 AMA Interim Meeting:
The following statement, recommended by the Council on Scientific Affairs, was adopted by the AMA House of Delegates as a Directive at the 2000 AMA Interim Meeting.
References 1. Notices. Federal Register: December 21, 1999;64(244):71440-71442. Also see AMA's Genetics and molecular medicine Web site |
Content provided by: CSAPH