CRISPR-Cas9 genome editing may not just be the greatest scientific breakthrough of this century, but also provides many lessons on patenting life science technologies.

The University of California announced late last month that their patent number 10,000,772 relating to Methods of RNA-directed targeted DNA modification had been granted by the USPTO. The patent covers the use of CRISPR-Cas9 technology in non-human organisms.

The CRISPR-Cas9 story has exploded in the last 5 years, with both industrial and academic institutions racing to secure patent protection in multiple countries. The highlights reel of this technology certainly raises eyebrows: its discovery is widely regarded as being worthy of a Nobel Prize, its ease of use accelerating genome editing in over two thousand research institutions across 61 countries. It is considered to be one of the most valuable biotechnology tools developed in recent memory, if not all time.

Huge Commercial Value

Genome editing is forecast to have a market value of USD$6.28 billion by the year 2022 although accurately assigning a dollar value to the overall market landscape can be difficult given there are no approved CRISPR products on the market as yet. Despite this, it is estimated that the value of being granted an exclusive licence to a key CRISPR-Cas9 cornerstone technology is between USD$100 million and $260 million. With that in mind it is easy to understand the magnitude and venom with which the commercial battle over the rights to licence CRISPR-Cas9 patents is being run. These patents are largely held by several academic institutions in the USA: the Broad Institute of MIT and Harvard and the University of California, Berkeley.

The battle lines are drawn around inventorship, and consequently, ownership.

Earlier in 2018, the European Patent Office (EPO) ruled that a fundamental European patent (2,771,468) by Broad was invalid arising from the fact that, what is often regarded as a formality – the naming of inventors – was incorrectly carried out. Being able to claim the December 2012 priority date for the ‘468 patent is crucial to Broad predating competing CRISPR patents filed just afterwards. The prolonged public relations battle between the main CRSIPR protagonists in the US has culminated in interference proceedings at the USPTO, and may yet affect other key European patents. This case shows how critical it is to clearly identify all inventors of an invention, especially one likely to have follow-on inventions. Identifying inventors can be a highly charged and political nightmare in fields in which collaboration between labs and across different institutions is commonplace.

Navigating ownership issues in Australia

The common law and IP Australia’s guidelines on the subject of IP ownership state that an invention conceived in the course of normal employment is the property of the employer. However, what constitutes ‘normal employment’ can be the stumbling block: this can differ between research institutions. In the case of co-ownership, the Patents Act specifies that the right to exercise a patent is granted to all patentees independently of each other, although the approval of all named owners is required to grant a licence.

Co-ownership of IP is rarely a good idea, but it can sometimes be difficult in an academic environment to determine what the alternative looks like. The IP Australia’s Toolkit for Collaboration has been created to help navigate the ownership issue. Best practice for scientists should include maintaining good laboratory notebook practice, such as clearly noting any novel significant findings in chronological order, and having this signed off regularly by a witness.

What next for CRISPR-Cas9?

The impact of CRISPR-Cas9 is reflected in the advances of genetic editing. Gao et al. describe a therapy involving CRISPR-mediated deletion of a pathogenic Tmc1 mutation, linked to early-onset deafness in humans. Removal of this single nucleotide mutation restored the growth of a proportion of inner ear cells, preventing development of hearing loss in adult mice. A similar approach is being employed for the treatment of sickle cell disease. Since the disease is caused by a mutation in β-globin, haematopoietic bone marrow stem cells are extracted from juvenile mice, and edited using CRISPR-Cas9 to either repair or replace the faulty gene before being reimplanted. CRISPR Therapeutics and Editas Medicine, the surrogate licensees or ‘spin-offs’ of UC Berkeley and Broad respectively, are looking to expand this research into clinical trials, although Editas Medicine apply a different approach. By disrupting functional repressors of fetal haemoglobin, they promote expression of a form of haemoglobin that does not contain the disrupted β-globin subunit, and thus reduce disease severity in animal models.

There are many unknowns to be unravelled before these exciting findings can be translated into life-saving therapies. The use of CRISPR-Cas9 genetic editing in transplanted haematopoietic stem cells carries risks. For example, the isolated stem cells might begin differentiating in culture before they are able to be used for treatment, rendering the prior genetic editing redundant. Or post-transplantation engraftment rates might be too low to confer any therapeutic benefit.

Jumping on the CRISPR-Cas9 train

The use of CRISPR-Cas9 technology for academic research in Australia is protected by the Patents Act which exempts acts for experimental purposes from patent infringement. But what if the experimental research results in commercial (and valuable) IP?

Currently, as CRISPR-Cas9 technology falls within the scope of multiple patent specifications, chances are multiple licences will be required to commercialise any new technologies which leverage the cornerstone technologies. The revocation of the Broad European ‘468 patent (and potentially more to follow) could significantly complicate matters, but nevertheless, it seems likely that companies wishing to license CRISPR technologies might have to acquire licenses from both Broad and UC Berkeley, possibly at significant cost. Innovation may be further hindered by the broad licences granted by both parties to their spin-offs in the development of human therapeutics, effectively blocking competition.

One way to address the multiparty approach to licencing CRISPR-Cas9 technologies may be the use of one or more patent pools. Widely utilised in telecoms and electronic technologies, patent pools enable streamlined patent licensing under terms generally designed to be Fair, Reasonable, and Non-Discriminatory to all parties. Discussions are currently underway for the development of a joint licensing pool coordinated by MPEG LA (of MPEG video codec fame). Under this model, companies wishing to licence key CRISPR technologies could do so non-exclusively, but patent pools also bring unique challenges to the table. Competition law issues aside, so far only Broad has submitted key CRISPR-Cas9 patents for consideration in the pool. UC Berkeley must also contribute foundational patents for it to be a commercial success.

Can patients be the winners?

In the end, if patients are to benefit from the CRISPR-Cas9 technologies, the question of where to draw the line in the IP sand has to be asked. If translational research is being hampered by a patent war, is it time to reassess the balance between legal monopoly and social responsibility? When polymerase chain reaction (PCR) technology gained traction in the 1980s and 1990s, Cetus, which was the patent owner, employed a “hands-off” approach, licensing the diagnostic rights to Kodak and Roche. These companies, along with manufacturers such as Perkin-Elmer, went on to develop further applications for PCR, now one of the most well-established biological research tools available today in widespread use for tremendous human benefit.

The PCR and CRISPR-Cas9 stories have some key differences: PCR was developed in a commercial environment and with a patent strategy in mind. The development of PCR won its inventor, Kary Mullis, the Nobel Prize in Chemistry in 1993. The significance of CRISPR-Cas9 is undeniable, but only time will tell whether ownership issues limit its evolution and universal adoption in scientific research.