United States District Court Judge Robert W. Sweet issued the 152-page decision, which invalidated seven patents related to the genes BRCA1 and BRCA2, whose mutations have been associated with cancer.
This is potentially huge. According to the Times, there are patents on about 20% of human genes. The scope of patents has liberalized over time, sometimes rightly–as with patents for computer algorithms–and sometimes wrongly–as, I would argue, with patents on genes.
As with all patents, the argument in support of gene patents is both technical and practical.
The technical argument is that cloning a gene is a considerable effort that makes it fundamentally different from the gene in vivo and also fundamentally more useful. The latter is a particularly powerful argument, since it is the same justification that was used to small biological molecules like adrenaline and Vitamin B12 and proteins like insulin.
The practical argument is, of course, that patent protection of genes creates an incentive for researchers to develop gene-based therapies and diagnostics. In the case of BRCA1 and BRCA2 this is of particularly importance, because these are markers for breast cancer.
The questions are 1) where the innovation is, 2) where the product/utility is, and 3) where the incentive is.
Where the innovation is
No one would argue that scientists or companies that clone genes are innovating at the sequence level. Genetic sequence innovation belongs to evolution. All that cloning does is to identify and extract a small bit of that information from the vast ocean of nucleic acids in a genome.
Particularly at the beginning, cloning was arduous and required lots of hard work, creativity, know-how, and luck. Cloning is not easy now (and still requires a lot of luck), but since we have sequenced the human genome, it is much, much simpler. Imagine you are Christopher Columbus sailing the ocean blue. Now imagine that someone hands you GPS. That’s the difference we’re talking about. GPS doesn’t make the seasickness any better, but it does mean you’re much more likely to get to India like you wanted.
Isolating DNA, purifying it, and cloning genes might sound fancy. But I am suspect about whether it is a strong enough basis for patents in the genomic era. I will cut it a little slack in my resolution, so read on.
Now one might argue that the real innovation is at the recognition that BRCA1 is a marker for cancer. I am not sure if this claim is made, but I am just trying to test out the possibilities. The problem with such an argument is that it means that our epidemiological data is for sale. Observations are not innovations, and patents ought to be for innovations.
Therefore, if the process of cloning a gene has any innovation, it is not at the level of the sequence and it is not at the level of the population. At best, it is at the level of the cloning method, and even this is shaky because a lot of these methods are standard protocols.
Where the product/utility is
The standard for biological patents seems to be that you make the substance more useful than it would be sitting in a chunk of flesh. I would argue that as the transition between flesh and pure substance becomes cheaper, easier, and more standardized, the potential utility of the flesh approaches that of the pure substance. This is particularly true for DNA, whose extraction and purification is facile.
More generally, the goal of a patent ought to be to spur and protect marketable discoveries. The gene itself is not a product and has no utility. It is rather a biological attribute of an individual, like an arm or a leg. Would you patent an arm or a leg?
Instead, the value derived from knowing something about the gene is what you do with that information. If someone develops a diagnostic, that is a real product with utility, a marketable discovery. It makes sense to patent the diagnostic; it makes less sense to patent the physical state it identifies or diagnoses.
Where the incentive is
This, of course, all brings us to incentive. As a society, we have little need to incentivize the collection of genes through patents. The incentives of basic research–grants, prestige, publications, degrees, positions, and tenure–are largely sufficient. What’s more, there is an indirect incentive on cloning genes when one allows patents that require that the genes have been cloned.
An example in the non-monetary scientific world is this: It is unlikely that Nature will publish my paper if I just clone a gene, but in order to get Nature-worthy results, I might have to clone the gene. Hence, the allure of Nature was sufficient to make the secondary, and necessary, contribution of the existence of the clone.
What we really want in society is to incentivize those marketable discoveries. These are the culmination of a lot of hard work and have clear financial potential following a set of fairly obvious steps. There is a complexity around giving patents for things in between the beginning and the end. Yes, it could be useful to provide a milestone patent so that the risk of undertaking a big project is not too large. Yet such a patent reduces the number of people who can take part in innovating on the second stage of that project. There is a barrier to participation for everyone besides the holder of the first patent, meaning that the chance of getting through that second part is smaller. Since the first part is not marketable by itself, society is left subsidizing risk with a dubious gain on reward.
The solution
A resolution to this contradiction is to distinguish between product and process patents. I learned about product and process patents when I was reading about patent policy in India. In order to enable generic companies to make cheap drugs, India used to recognize process patents, but not product patents. (Here is a nice article from Kannan Sivaprakasam on the transition from process to product patents in India in 2005.) Therefore, if you made a small molecule, you did not own the molecule, but rather the method to synthesize it. If someone else developed a different method to synthesize it, he could get a patent for that; while he did not contribute the molecule, he may have contributed a cheaper way to make it–another marketable discovery.
I see the problems with this, because having the molecule itself is of value and we want to incentivize those who are first to market with new ideas. Therefore, I think a hybrid system would be best, where we sharply reduce the length of product patents but increase the length of process patents. This way, we shield the innovation of the first to market, but provide incentives for process innovation sooner.
This could have an impact on the gene therapy and diagnostic market, but not by considering genes patentable. Genes are not products and they are not discoveries. Perhaps the specific methods used to clone them, like the specific arrangement of 1′s and 0′s in an algorithm, are patent-worthy. After all, it is this difficult cloning process that supposedly justifies patents on the stuff that makes us who we are.
I think we would be better off if we kept the method to ourselves and left the substance to Nature.
