One of the major battles going on in patent law is that concerning the CRISPR gene-editing technology. Basically, this technology allows for the modification of genes in living organisms. As one can imagine, if this technology could be perfected for use in human DNA, the scientific and business possibilities are vast and valuable. Involving this technology is a patent battle between the University of California Berkeley and the Broad Institute, which includes Harvard and MIT. The battle is one for interference: Berkeley was the first to publish on the use of such technology on prokaryotes. The Broad Institute’s patents involved the use of the technology on eukaryotes, which Berkeley claims was obvious. The interference proceeding was decided on February 15, 2017, by the PTAB, at which it ruled that there was no interference as Broad’s work was considered not obvious. The Broad patents are thus valid, and Berkeley’s, which is still undergoing examination, would likely be as well. As the patents become confirmed, a larger problem looms: the impact of these patents on innovation concerning CRISPR gene-editing technologies.
In a talk given at NYU Law School on March 23, 2017, called The CRISPR Patent Battle: Implications for Downstream Innovation in Gene Editing, a panel of experts in the field discussed the impact of the ruling on innovation. The talk was moderated by Prof. Anne Hassett of NYU Law, and the focus was on a paper by Prof. Jorge Contreras of the College of Law at University of Utah and Prof. Jacob Sherkow of New York Law School, called CRISPR, Surrogate Licensing, and Scientific Discovery. The paper focuses on the ongoing commercial enterprises that had already been entered into by the institutions involved and the focus of their business ventures. The model of licensing used by the institutions is the use of a surrogate company. These surrogate companies then can decide on granting non-exclusive and exclusive licenses to other companies on specific fields of use. The surrogate model with respect to CRISPR, compiled by the authors of the article, is presented below:
This surrogate model is beneficial for universities as it maximizes profits while minimizing risk on them. As the article states, this model “permits the university to focus on a broader range of commercialization projected…and delegates the job of licensing to experts focused on the relevant technology.” The universities get a share of the profits from these licenses. However, there is a major negative to innovation itself with this model. Each surrogate had been granted exclusive rights to use CRISPR to develop human therapeutics for any and all of the genes of the human genome. With such overly broad exclusive rights, a bottleneck is created.
As we see from the chart above, there are only five surrogate companies with exclusive licenses on human therapeutics. Yet, there are close to 20,000 genes in the human genome. For any single gene of interest, the use of CRISPR to develop therapeutics with them would require a sublicense with at least one of these surrogate companies. Given the ruling that Broad’s patent is valid, if Berkeley’s patent application is also granted, it may require any institution working on human therapeutics to get a sublicense for both sets of patents to ensure they are not infringing. The problem is thus the large number of possible genes to explore and possible fields of use, but only five surrogate companies to decide on what sublicenses would be granted and to whom. This creates two issues. First, it is unlikely that the surrogate companies are pursuing work on every single gene in the human genome. The surrogates would likely only be looking to grant sublicenses to companies who are working on specific genes that the surrogates find of interest. This would lead to sublicenses to a select few, but other genes in the genome, of which these surrogates hold exclusive licenses to, may sit unclaimed and ignored. Even if another institution has an interest in exploring those genes, the surrogates’ lack of expertise for that gene could lead to a slower consideration of that institution’s proposal.
A second problem concerns the sublicenses themselves. The Contreras/Sherkow article points to an exclusive license between Editas Medicine, a surrogate, and Juno Therapeutics to illustrate this issue. The license grants Juno the use of CRISP to develop therapies across multiple genes. By providing a sublicensee multiple targeted genes, no other company can pursue any research on those genes using CRISPR, nor can they attempt to get a license from Editas due to Juno’s exclusivity. This leads to a second bottleneck to innovation: Juno may have gotten a license to research on all those genes, but there is no guarantee that they would work on them. It is quite possible that Juno would focus on just a select few for years, leaving the other genes that had been marked by both Juno and Editas to have innovative benefits to be placed aside until Juno can get to it.
Other than the bottlenecking at the surrogate level, a further problem too concerns a focus on profit. Since the surrogates are for-profit companies, the focus of their licensing work would be on genes that would maximize profit. This leaves unprofitable therapeutics as afterthoughts. However, there is likely a set of genes which have socially valuable benefits. It is likely that therapeutics affecting “[r]are diseases, or those that disproportionately affect disadvantaged populations or regions” would be put to end of the queue.
With the possibility that both sets of patents will be valid, all these problems could be multiplied, as companies may need to get licenses from both Broad’s and Berkeley’s surrogates for work on the same genes in order to use the CRISPR technology. It could also be possible that if the surrogates grant licenses on the same genes to different companies, a full innovation stop could be hit on those genes as each of the sublicensees could be infringing on the patent of the other patent-holder. With such issues, it is imperative that these licensing agreements be reconsidered. The Contreras/Sherkow article proposes a few ideas that would be ideal. One consideration is for Broad and Berkeley to consider reserving the right to license certain areas that CRISPR could be of use for that the surrogates would not be pursuing, based on their lack of expertise or interest. This would allow for a broader range of research to be conducted with the use of CRISPR and would still allow the patent-holders to maximize their profits. Another idea would be more regulatory; the NIH could look at platform technologies the same way as it does with research tools. As part of the Nine Points core licensing values that most research institutions follow, one of the points is that patented research tools must be as broadly available as possible. To consider that as a key part of being a research institution could push the patent-holders to modify their agreements. There is lastly just a general hope that, given that the research performed in creating this technology is publicly funded, the patent-holders would be able to notice that negative impact on research of the surrogate model being used, and consider modifying it to pay back to the public what it invested in. Broad and Berkeley can still focus on profiting from their research, but the hope is that furthering innovation could share the same importance to them in their licensing arrangements as money.
Jun Tong is a J.D. candidate, ’18, at the NYU School of Law.