Are there any circumstances in which a private citizen, a public institute or a corporation can legitimately hold a patent on a human gene? All patent laws refuse protection to discoveries about the natural world; Columbus could not have patented America. But genetics has become so competitive and potentially profitable that researchers (or, more often, their corporate sponsors) are nibbling away at the spirit, if not the letter, of the law. The distinction between discovery and invention is becoming blurred.
There was a spectacular illustration of this four years ago, when the US National Institutes of Health, on behalf of its employee Craig Venter, applied for patent protection for some hundreds of pieces of DNA that were almost certainly identical with authentic "ends" of human genes. Some were recognisable as the ends of genes already known. Others were the ends of genes of unknown function. There was a great outcry from the research community. People protested that it should not be possible to patent pieces of DNA whose useful purpose could be determined only by further research. Both the British and the French governments put out statements deploring this development, while the Medical Research Council in Britain sought to protect its position by applying for patents for data similar to Venter's. Eventually the US patent applications were refused on the grounds that the invention's utility had not been proved.
But the research community remains alarmed. Last year, the American Society of Cell Biologists called for a change in the patent laws to deal with what it saw as growing inequity between researchers. Patents in the field of genetics rest on a mass of research which has appeared in the scientific literature; but patents are awarded only to the handful of people who take the decisive-and first patentable-step. If patent law cannot easily be modified to meet the cell biologists' wish that the rewards be more equitably shared, then the remedy must lie with academic institutions.
These issues have become prominent because the techniques for analysing genes (and pieces of them) have been made marvellously powerful in the past few years. Genes are lengths of DNA strung together in the human genome, and are exactly defined by the sequence in which chemical units called nucleotides (of which there are four) are linked with each other. The practical value of this information is, for the time being, mostly in the diagnosis of genetic diseases, although treatment for some of them may not be far away. Those who inherit these conditions, which range from haemophilia to some diabetes, have inherited from one or other of their parents (but sometimes from both) a version of the gene which differs from the disease-free variety.
At the very least, a researcher who knows the sequence of a disease-linked gene has a potential diagnostic test to put on the market. He, she or the sponsor will, in the present climate, wonder whether the cost of seeking patent protection would be worthwhile. The initial cost may be high (perhaps $50,000 worldwide), but who can say that the financial benefits will not be much greater, especially when a patent-holder may enjoy a sellers' market and the freedom to charge high royalties for the use of the invention?
But is a genetic test really an invention? Even the altered genes occur naturally, so their identification could be counted as a discovery, like Columbus's journey to the New World. Not so, says patent law. For what the inventor and the sponsor sell to their customers is a recipe for telling whether or not the gene in question is altered in the individual to whom the test is applied. Often the inventor will supply a "kit" which may include two pieces of naturally occurring (and unaltered) DNA which can be used to isolate the regions of a genome in which the disease-linked alterations occur. In other words, the successful gene hunter is providing more than mere knowledge.
If this is invention it is rarely a solo effort. The search for the altered genes involved in family-inherited breast cancer illustrates the point. Several research groups in Europe and North America have been engaged in this task for many years, in polite but intense competition with each other. Each group has involved doctors with access to families in which breast (or ovarian) cancer seems to recur.
Last year a research group at the University of Utah announced the identification of a gene now called BCAR-1 ("BC" stands for "breast cancer") which seemed to be altered in many but not all cases of familial breast cancer. The normal function of the gene is still not known, although it seems to be involved with regulating such occurrences as the replication of the DNA in a cell in preparation for its natural division into two. The presumption is that an alteration of the gene can let cell division run amok, causing tumours.
The disappointment for the Utah group was that, among the families to which they had access, more than a dozen distinct alterations of BCAR-1 were associated with the predisposition to breast cancer. That meant that no single recipe would make a saleable testing kit. Undeterred, the Utah group continued to collaborate with its chosen commercial partner Myriad, a spin-off from the university.
The Utah group recognised that alterations of the BCAR-1 gene could not be the sole cause of familial breast cancer, and the hunt was intensified for a second gene, called BCAR-2. Towards the end of last year, a joint effort by the Cancer Research Campaign and the Sanger Centre in Cambridge, UK (jointly supported by the Medical Research Council and the Wellcome Trust), announced the full characterisation of this gene, declaring its belief that the information it had acquired should be in the public domain. True to its word, it made the full sequence of the gene available on the internet.
Then came a surprise. The Utah/Myriad group let it be known that it had already done the same work and had applied for a patent, which would pre-date any application from the British team. What the outcome of that will be is anybody's guess. But the incident is a vivid proof that the assignment of rewards for success is often a matter of pot luck.
Worse than that has happened. Craig Venter left the National Institutes of Health to found The Institute of Genomic Research (TIGR), a not-for-profit organisation under US law, at which he has further developed his flair for the automated analysis of human DNA and for using computers to make sense of the huge amounts of data recovered. TIGR has made a deal with a commercial corporation called Industrial Genome Sciences (IGS).
There was a great rumpus at the end of 1994, when TIGR and IGS made it known that they would make their vast amount of data available to other researchers, but only on terms requiring outsiders to give IGS an option on commercially usable results arising from research subsequently carried out. One measure of the potential benefits is that Smith-Klein Beecham, the pharmaceutical company, has paid $150m to IGS for a first option on its data.
A curious twist has now arisen. A few weeks ago, IGS announced that it had obtained the complete genetic sequence of the bacterium Staphylococcus aureus, the common organism whose virulent strains are creating problems in hospitals, but that the sequence would not be made public. IGS is probably already embarked on the search for drugs to prevent the replication of these dangerous organisms. Meanwhile the genetic sequence-of great importance for those who wish to understand the emergence of virulent bacterial strains-remains private.
Several conflicts of interest have thus arisen. Researchers are in conflict with others in the rush to a patentable result, with all the damage that does to the civility (and the effectiveness) of the research profession. And to the extent that researchers have pre-committed the exploitation of patents to particular companies (in whose founding they have usually played a part), the impartiality of their institutions, as sources of knowledge for the community at large, is compromised.
But the most serious conflict is between the public and the private interest, typified by the secrecy of the commercial restrictions attached to much of the data gathered in the private sector. Academic researchers make it a point of honour that genetic data accompanying published claims should be made available through the databanks, but commercial researchers have no such obligation. IGS's Staphylococcus sequence, for example, will presumably remain secret until it has been milked dry.
Patent law could be changed to blunt the edge of some of these conflicts. In Europe and Japan, prior disclosure is a bar to the subsequent granting of a patent, but in the US there is a one year period of grace between publication and the deadline for a patent application. US practice avoids the withholding of information while the patent lawyers scratch their heads, and is thus preferable. The publication of information about patent applications also needs attention-with the world's patent offices at their wits' ends to keep up with the flood of innovations in fields such as genetics.
Whether the rash of diagnostic genetic tests brought forward in the past few years should be patentable at all remains the crucial question. If somebody has identified a gene and the alterations of it link to a disease, devising a genetic test is a fairly obvious business. But all patent laws include as a test of originality that innovations should not be "obvious."
Commercial organisations would protest that such an interpretation of the existing laws would rob them of the incentive to seek innovations. But that would be misleading. The long term incentive to search for altered disease-linked genes is that it will point to ways in which the malign effects of the alterations can be nullified by drugs or other means. That is where the real riches lie for the pharmaceutical industry. Squabbling in the meantime over patents for genetic tests is unseemly-and against the public interest.
