From HMS BEAGLE
Genes
Is It Necessary and Is It Evil?
Posted March 17, 2000
by Richard F. Harris
This article also appears in Current Biology.
--------------------------------------------------------------------------------
Last October, biologists' neck hair rose when J. Craig Venter announced his company, Celera, had filed 6,500 provisional patent applications for human genes. Henry Ford mass-produced automobiles - it seems evident we are now entering an era in which intellectual property is rolling off the assembly lines. Is this really the ultimate legacy of Watson and Crick's elegant double helix? And what does it portend for the future of biology?
On April 22, 1996, Europe's patent office received an application from a U.S. company called Human Genome Sciences, claiming the rights to the entire genome of Haemophilus influenzae. This patent also asserts the right to that information in any computer-readable form - effectively preventing it from being posted in any public database for any scientific use. Scientists wanting to interpret this genome and compare it to others would have to resort to paper and pencil - or presumably pay for access to a private database.
Each of these cases is sobering in its own way, and each raises distinctly different issues. Consider first the attempt to patent the entire Haemophilus genome. Venter's name is on that patent - and his wife's is on a very similar application that would cover the whole genome of Mycoplasma genitalium. These patents have not been issued, and it's not clear they will ever be. But the filing of the patents cuts to the heart of an issue surrounding the patenting of genes: just how much biology should be turned over to the private ownership of individuals and companies?
The concept of patenting has been around for centuries. The original idea arose in England, where patents were first used as a means to lure printing presses into the country, in exchange for a limited monopoly on the printing technology. Later, patents were used to encourage guilds to reveal their trade secrets; a monopoly right was granted in exchange for information that others could then use to improve practices in science and the practical arts. Indeed, few would dispute that patents have encouraged innovation. But some gene patents are now testing the limits of the patent system.
There is clearly a place for gene patents, when the gene in question is well characterized, and when there is a clear use for the information. Patents, after all, must satisfy three basic criteria: the invention must be novel, it must not be obvious, and it must be useful. Michael Morgan, at the Wellcome Trust, helped craft the most restrictive (albeit voluntary) guidelines for gene patenting - the "Bermuda Agreement," which stipulates that the raw data from the Human Genome Project will be placed immediately in the public domain and therefore not subject to patenting. He and his U.S. colleague Francis Collins (head of the Human Genome Project) agree that it is appropriate in some circumstances - such as when there's a clear, immediate use - to patent genes. Collins is even the "inventor" of the genes for cystic fibrosis and neurofibromatosis, to name two.
The sticky bit comes, however, when a claim lacks an obvious use. The U.S. Patent and Trademark Office has approved a few patents for mere gene fragments - where the "utility" is essentially as scientific probes to find related genes.
It's one thing to patent tools. Shouldn't the inventor of the electron microscope be granted some consideration for his creativity? Likewise, shouldn't the "inventor" (that's the word the patent offices use) of restriction enzymes be granted some reward - in exchange for telling the world exactly what the invention is and how it works? But some of these new gene patents have an insidious nature to them. True, they are tools that can be used to discover more genes. But in this case, inventors can assert that the new discoveries are, in effect, an extension of their tools - therefore also part of their intellectual property. It's as though the inventor of the electron microscope could also claim all the discoveries made with that instrument.
Table 1. Gene patents issued by the U.S. Patent and Trademark Office as of year-end 1999.
Holder No. of patents
U.S. government 388
Incyte Pharmaceuticals 356
University of California 265
SmithKline Beecham 197
Genentech 175
Eli Lilly 145
Novo Nordisk 142
Chiron 129
American Home Products 117
Isis Pharmaceuticals 108
Mass. General Hospital 108
Human Genome Sciences 104
University of Texas 103
Institut Pasteur 101
Source: PricewaterhouseCoopers. Patents typically take three years to issue. European patents are revealed 18 months after they are submitted and while they are still pending. The U.S. is adopting to a similar system. Pending patents are not included in the above chart.
Collins has argued that the U.S. Patent and Trademark Office has been too lax in granting patents on the basis of their utility as research tools (see table 1). The patent office reconsidered its utility guidelines for genes and tightened them somewhat in December 1999. By and large, gene fragments are no longer considered useful (or patentable). But if an inventor can identify a gene, a gene product, and state a use, the office will issue a patent. And useful in this case doesn't have to mean potentially therapeutic or valuable in the laboratory. A "use" is simply describing how the gene product functions in nature which - from a practical standpoint - is to say not necessarily useful at all. These revised guidelines will not have much of an effect on the 10,000 or so gene patents now pending in the U.S. Patent office officials figure only about 500 patents will be rejected in their current form.
Patents typically take three years to issue. European patents are revealed 18 months after they are submitted and while they are still pending. The U.S. is adopting to a similar system. Pending patents are not included in the above chart.
Which brings us back to Henry Ford. A few companies - notably Human Genome Sciences, Celera, and Incyte Pharmaceuticals - claim to have thousands of patents pending. Some are mere placeholders - Celera's "provisional" patent applications expire 12 months after they are issued, so the company will need to pick and choose which are actually worth the trouble of fleshing out and filing in earnest. Venter asserts he will only hold a few hundred patents on human genes when all is said and done.
But this patent gold rush has disturbing consequences for biologists. Part of it is uncertainty about how these patents could complicate their research. Academic freedom does not extend to using other people's intellectual property - a patent holder has a right to exclude anyone from using his or her invention. Historically, academics have not been prevented from using patented genes - indeed, companies often have a strong financial incentive to let others discover more about their patented material because it increases the value of the patent. But the moment money changes hands, say at an academic genetics lab, patent holders are quick to step in and assert their rights.
And it is ironic that in some of the most celebrated cases regarding gene patents, the villains turn out to be academic institutions. Allen Roses, now senior vice president of genetics research at Glaxo SmithKline, discovered that people who carry the ApoE4 allele are at greater risk for developing Alzheimer's disease. Duke University, where Roses made that discovery, licensed that patented discovery exclusively to a small biotechnology company, which is now trying to turn a profit by marketing that test and others like it. That galls the folks in academic genetics labs, who were told to stop offering their own versions of the ApoE4 test (at considerably lower cost).
William Haseltine, chairman and CEO of Human Genome Sciences, argues the whole patent issue has been overblown. He says, for example, more than 1,000 patents have been issued in connection with HIV, yet research on AIDS has hardly been choked. Indeed, Haseltine says the U.S. government holds a patent on what amounts to the whole AIDS virus - that's the basis of the ELISA assay that was hotly contested with the Institut Pasteur.
But there is a deeper issue here. Academic researchers who used to labor for years to discover and understand the basics of a single gene are aghast to discover that intellectual property can now be manufactured by the bushel barrel. Mostly what it requires is some DNA sequencers, a cadre of Ph.D.s, and a computer algorithm that can spot homologies between novel stretches of DNA and sequences of known function. Presto, a gene patent is born, and is the exclusive provenance of its owner for at least 17 years.
Necessity is not the mother of these inventions - indeed, the proud parents usually intend to put them up on the auction block as quickly as possible. And many orphans are likely to languish. The other part of the social contract - public disclosure - is more frequently taking place in patent filings rather than the pages of peer-reviewed journals. Patent office Web sites are rapidly becoming an internet destination to rival PubMed.
It seems inevitable now that most useful human genes will be covered by patents. But even the big drug companies have found a limit to the gene-patenting craze. Ten of the pharmaceutical giants have joined with the Wellcome Trust to collaborate in finding hundreds of thousands of single-base mutations in DNA (single nucleotide polymorphisms or SNPs). They will try to patent this SNP data, as it could be a valuable tool for discovering genetic variations that correlate with common diseases. But they have decided that they will not enforce the patents (if, indeed, they receive them). The patents would simply prevent others from locking up this information for their private use. In this case, at least, they have decided that genetic information best serves its masters by being free.
Richard F. Harris is a science correspondent at National Public Radio and past president of the National Association of Science Writers.
-------------------------------------------------------------------------------- |
