Barriers to commercialisation

18.9 As discussed elsewhere in this Report, a primary purpose of patent laws is to provide an incentive for innovation. Intellectual property rights, and patent rights in particular, are attractive to firms because they create the prospect of charging others monopoly prices for access to their intellectual capital and prevent others (‘free riders’) from taking advantage of their investment. However, inadequate intellectual property protection and management has been identified as one of the major barriers to commercialisation in the Australian biotechnology sector.^[3]

18.10 Dr Dianne Nicol and Jane Nielsen have argued that ‘the regimes protecting IPRs [intellectual property rights] may prove to be a significant barrier for the development of the Australian industry’.^[4] They suggest that the patent system is ‘crucial to the biotechnology industry in order to reward and encourage innovation … [but] it is becoming apparent that the same regime may hinder the research efforts of Australian companies by restricting access to research tools and technologies’.^[5]

18.11 As noted above, patents may also act as a barrier to research and a disincentive to commercialisation. The problems cited in Chapter 12 are generally as relevant to product development as they are to further research. In particular, as Nicol and Nielsen suggest, biotechnology companies face unique challenges due to: the research-intensive nature of the industry; the massive increase in patent activity in the area of biotechnology; the preponderance of upstream patents with broad claims; and the reliance of downstream companies on access to patented research tools and techniques.^[6]

Patent thickets

18.12 ‘Patent thickets’ are a consequence of multiple upstream patents.^[7] A patent thicket has been described as ‘a dense web of overlapping intellectual property rights that a company must hack its way through in order to actually commercialize new technology’.^[8]

18.13 The existence of multiple patents has also been described as the ‘tragedy of the anti-commons’; namely, the under-use of a scarce resource, where multiple owners exclude others and no one has an effective privilege to use the resource.^[9] The proliferation of gene patents, including multiple patents on research tools, may impede downstream research and innovation if access to tools required for further development is difficult to negotiate. Nicol and Nielsen suggest that in extreme cases projects may even be abandoned. They argue that:

If negotiations are required to be undertaken with a number of parties, the risk of negotiation breakdown is increased … Depending on the stage at which breakdown occurs, this may mean that projects are either not commenced or are abandoned at some stage into the research process … As the number of relevant intellectual property rights increases, the task of inventing around becomes more onerous, and project abandonment may become inevitable.^[10]

18.14 Patent thickets may also raise production costs. Even if multiple licences are successfully negotiated, the cost of obtaining numerous licences may add to the cost of development and may also be passed on to consumers in the form of higher prices.^[11] These increased production costs may affect commercial incentives for pursuing downstream product development and marketing.

18.15 The issue is not confined to gene patents, but it arises in relation to gene patents because different patents over the same gene may contain overlapping claims. A gene contains coding DNA sequences (exons), non-coding regulatory DNA sequences, and functionless introns.^[12] Conceivably, separate patent claims might be made on each of the exons as expressed gene fragments; another claim could be made over the complete expressed sequence; another on a promoter sequence; and others over mutations known to have the potential to cause diseases. Patent thickets could present a problem in this area, for example in the development of genetic diagnostic tests or therapeutic proteins, where access is required to genetic information covered by multiple patents. Professors Michael Heller and Rebecca Eisenberg have expressed concern that ‘a patent anti-commons could prove more intractable in biomedical research than in other settings’.^[13]

18.16 One of the questions Nicol and Nielsen addressed in their empirical study of patents and the medical biotechnology industry (Nicol–Nielsen Study) was whether an anti-commons had emerged in Australia.^[14] They reported that respondents to the Study did not describe significant problems with the enforcement of multiple research tool patents:

In part this is because a number of the most aggressively enforced research tool patents do not exist in Australia, or, if they do exist, they do not appear to be enforced. However, we expect that these or other patents may well be enforced in the future. Hence, it would be premature to say that the Australian industry is free from the rigors of research tool patent enforcement.^[15]

18.17 Nicol and Nielsen also reported that, although the Australian patent landscape is becoming increasingly complex, the number of problematic patents affecting research is quite small. They suggested that:

in part the reason for this is that if there is a higher level of encumbrance research will be redirected. We are unable to state with any level of precision the number of research projects that are abandoned because there are too many problematic patents in the area. However, we know that this problem does exist.^[16]

18.18 On balance, Nicol and Nielsen concluded that their results did not provide conclusive evidence of either the existence or absence of an anti-commons in Australia, although they did note the potential for one to develop due to ‘ongoing increases in the number of patents, more vigilant enforcement and the increasing complexity of research paths’.^[17]

Royalty stacking

18.19 Royalty stacking is a problem caused by a multiplicity of overlapping patents, especially over upstream products. The need to pay multiple licence fees may force up prices and discourage innovation and product development. In the context of pharmaceutical patents, Phillip Grubb suggests royalty payments for the use of research tools may be problematic because:

it will often be the case that a number of different tools or technologies have contributed to the drug development, and whereas a single royalty of one or two per cent may be an acceptable burden, an accumulation of such royalties soon adds up to an unacceptable amount.^[18]

18.20 The Organisation for Economic Co-operation and Development Report, Genetic Inventions, Intellectual Property Rights and Licensing Practices: Evidence and Policies (OECD Report), linked concerns about patents over research tools with the problem of patent thickets and royalty stacking and suggested that together these have the potential to raise the costs of conducting research and ultimately the costs of products. The OECD Report suggested that royalties could comprise up to 20% of the net price of some products.^[19]

18.21 Often, however, companies are able to address stacking problems by contractual solutions, for example through placing limits on cumulative royalties or requiring up front payments rather than royalties.^[20] The OECD Report noted that contractual solutions are generally pursued because it is in the interests of companies to accommodate reduced royalties to enable agreements to be made for the use of patents. It suggested that biotechnology projects that require patented technology to be licensed-in rarely fail due to royalty stacking concerns.^[21]

18.22 It is unclear whether royalty stacking is a serious problem for the Australian biotechnology industry. However, given that the industry is largely comprised of upstream companies, it may be a lesser problem here than in industries overseas with a more significant downstream component. Respondents to the Nicol–Nielsen Study reported mixed experiences of royalty stacking. One intermediate company stated that they had not encountered it, while one upstream company predicted that ‘in the future, when conducting licensing negotiations, companies may well be exposed to licence stacking and overlapping royalty structures’. In addition, mixed views were expressed about the reactions of downstream pharmaceutical companies to royalty stacking:

One respondent said that large pharmaceutical companies abhor royalty stacking. However, a pharmaceutical company respondent noted that although reach-through royalties and divided ownership don’t help in the drug development process, ‘they are not showstoppers’.^[22]

18.23 Patent pools^[23] are a mechanism for overcoming some of the difficulties of access to research tools and technologies caused by a multiplicity of patents.^[24] Commercial products such as therapeutic proteins or diagnostic genetic tests are likely to require access to many gene fragments; a bundle of licences collected in a single licence arrangement can overcome the problem of dealing with multiple patent holders or licensees.^[25] Patent pools are discussed further in Chapter 22. Patent pools also raise competition issues, which are discussed in Chapter 24.

Broad patents

18.24 Chapter 12 described broad patents as patents that grant broad rights to the patent holder and which may be seen as covering applications invented later by someone else. As noted in that chapter, such patents may discourage further research and innovation because researchers may be concerned about infringing them, or because of the potential cost of licence fees associated with the use of the patented invention. These concerns are also relevant to industry, both as constraints on further research, and more particularly, because potentially the cost of licence fees may decrease the returns on products developed using the patented technology. Companies may then attempt to offset this decrease through charging higher prices for products.

18.25 The flow on effect of these problems may include:

• increases in the cost of healthcare products;

• fewer products may be available if the development of some products is abandoned; and

• inefficient use of resources due to paying licence fees or ‘inventing around’ unnecessarily.^[26]

18.26 The Nicol–Nielsen Study reported that 12 of the 49 company respondents to the survey believed the grant of broad patents had an inhibitory effect on their research. Respondents also noted that, despite this, they continued to seek patents that were as broad as possible.^[27] However, some respondents commented that for some companies, ‘inventing around’ might be a workable strategy for dealing with broad patents. In particular, respondents from the pharmaceutical sector ‘were generally of the view that it is not possible to obtain broad patents that block research in the pharmaceutical industry because of the ability of researchers to invent around’.^[28]

18.27 Some concern about the stifling effect of broad patents was also expressed in submissions.^[29] However, one submission noted that overly broad patents are unlikely to be enforceable.^[30] Others commented that there is little clear evidence that research is being stifled or that such patents are adversely affecting the Australian biotechnology industry.^[31]

18.28 Broad claims are often a feature of patents granted in the early stage of a new technology. Initial developments in a field often underpin a range of subsequent work and, as such, represent significant contributions that have multiple applications.

18.29 The ALRC has developed recommendations that may help to address some of the concerns about inappropriately broad patent claims. These include amending the Patents Act 1990 (Cth) (Patents Act) to include ‘usefulness’ as a requirement in the examination of patent applications; and to require patent examiners to be satisfied on the balance of probabilities when assessing all the requirements for patentability that are relevant at the stage of examination.^[32]

18.30 In addition, Chapter 8 makes recommendations directed at improving the education and training of patent examiners and developing guidelines to assist examiners in applying the requirements for patentability to inventions involving genetic materials and technologies.^[33]

Reach-through provisions

18.31 Chapter 12 discusses the problem of reach-through provisions in licence agreements in the context of research. This section discusses the implications of reach-through provisions for the biotechnology industry.

18.32 Reach-through provisions in a licence agreement grant a patent holder future rights in new products that might result from the use of a licensed patent. Reach-through provisions may include the right to own or license the intellectual property in future products. Reach-through provisions are most common when the holder of a patent over an upstream technology licenses it to other companies further downstream. They may also be included in materials transfer agreements.^[34] Reach-through provisions, therefore, effectively give the patent holder what Heller and Eisenberg have described as ‘a continuing right to be present at the bargaining table as a research project moves downstream toward future product development’.^[35]

18.33 Reach-through provisions may deter investment in developing technology if it is unclear whether the technology can be freely exploited. They may also deter future development by restricting the rights of the licensee to exploit new technology that results from working with the patented technology. This may be particularly problematic if a number of reach-through rights are stacked on downstream technologies.

18.34 The Nicol–Nielsen Study suggested that reach-through provisions are frequently included in licensing agreements entered into by participants in the Australian biotechnology sector and appear to be problematic in a large number of negotiations.^[36] Heller and Eisenberg suggest that reach-through provisions may contribute to the emergence of an anti-commons.^[37] The Nicol–Nielsen Study did not hear complaints about stacking of reach-through provisions, however, it noted that this did not mean problems of this kind did not exist.^[38]

18.35 Submissions expressed some concern about reach-through provisions, although this concern was often focused on reach-through patent claims.^[39] Whereas reach-through provisions are conditions on the use of a licence of a patent, reach-through patent claims seek to claim rights to a future invention on the basis of a currently disclosed invention, thereby extending the scope of the patent holder’s monopoly.^[40] Reach-through patent claims are discussed further in Chapter 6.

Blocking patents

18.36 Broadly defined, blocking patents are patents which stifle developments by others. They may occur where one patent holder has a broad patent over an invention (a dominant patent), and another patent holder has a narrower patent over an improvement to that invention or a new invention that relies on access to the original invention (a dependent patent).^[41] The holder of a dependent patent will be precluded from practising the improved invention unless it can obtain a licence over the dominant patent. Similarly, the dominant patent holder may not exploit the improved invention without a licence from the dependent patent holder.^[42]

18.37 Blocking patents may cover broad or foundational technology—a patent over a gene sequence is an example—which is not exploited or licensed, thereby blocking others from using the technology. The effect of a broad blocking patent may be to prevent whole areas of research, particularly where the patent holder chooses not to exploit the patent themselves. In such cases, or where the patent is foundational to other research, the capacity for others to undertake further research may be curtailed and the benefits from the technology may not flow to the public.^[43]

18.38 The value of a patent is highly dependent on the patent holder’s ability to exploit it.^[44] Patents that block a patent holder from exploiting their patent may devalue the patent, and consequently affect the holder’s ability to attract investment.

18.39 The existence and effects of blocking patents were examined in the Nicol–Nielsen Study. The Study found that a significant number of respondents regarded blocking patents as a real issue in the biotechnology industry, although many commented that they could not see the value of companies obtaining patents purely for blocking or defensive purposes.^[45] However, Nicol and Nielsen commented:

many respondents who participated in interviews either had patents that they did not currently exploit, or knew of companies who did not currently exploit. In many cases, these patents were not licensed or otherwise transferred, although this may have been due to a number of reasons.^[46]

18.40 The Nicol–Nielsen Study also found that several respondents to the company survey had altered their research program due to a patent blocking access to research tools or materials.^[47] Some respondents commented that they avoided areas of research where they did not think they would be able to get access to necessary technology due to the presence of blocking patents. Others changed the direction of research where they found themselves blocked, or invented around the patented technology.^[48]

18.41 Despite the reported concerns, responses suggested that successfully negotiating a licence could solve issues created by potentially blocking patents, and one respondent commented that research is blocked only in exceptional cases.^[49] However, Nicol and Nielsen observed that in many instances, respondents did not even try to negotiate as they regarded the patent holder as unlikely to negotiate.^[50]

18.42 Another mechanism for dealing with blocking patents is compulsory licensing, where the holder of the patent can be required to license the technology to allow others in the industry to exploit it, or to practise their own patents. The compulsory licensing provisions in the Patents Act are discussed in Chapter 27. There may also be competition issues, which are discussed in Chapter 24.

Dependency and uncertainty

18.43 As discussed above, a dependent patent is a patent on an invention, the exploitation of which is prevented by an earlier patent. The OECD Report suggested that the rapid proliferation of gene patents could cause commercial uncertainty and cited the example of different patents for inventions claiming ‘a partial gene sequence (for example, an EST), the full-length cDNA or gene, and the protein encoded’^[51] leading to uncertainty about which patent holder would be able to prevent the others from using the later invention. The OECD Report stated that ‘while licensing under uncertainty about the extent of property rights is not new to the pharmaceutical industry, too much litigation could again slow progress, raise end-product costs or discourage entry to certain fields of enquiry’.^[52]

18.44 The OECD Report further noted that:

While official statistics show that the number of patent applications and grants is on the rise, little is known about who is licensing what technologies to whom and under what conditions. Firms claim that it is increasingly difficult to assess whether they have ‘freedom to use’ their own in-house or licensed technologies as the web of patents becomes more complex and overlapping.^[53]

18.45 Compulsory licences may offer a solution where the holder of a dependent patent is unable to obtain a licence over the dominant patent. Chapter 27 discusses the way in which the compulsory licence provisions under the Patents Act apply to dependent patents.

Refusals to license

18.46 As discussed in Chapter 22, licensing is a means by which rights in patented technology may be transferred.^[54] There are two main types of licences:

• those where a party needs to acquire the rights to use a patent in order to do further research or development (licence-in); and

• those where the right to use patented technology is granted by a patent holder to another party to allow further research, the development of a new product, or the exploitation of a product (licence-out).

18.47 Chapter 16 noted that the level of licensing in the Australian biotechnology sector is ‘prolific’ but it also noted a finding by Ernst & Young that more than 20% of firms surveyed reported abandoning a project because of an inability to obtain a licence.^[55] However, only six of the 49 companies that responded to the Nicol–Nielsen Study reported being refused licenses.^[56] Interview responses reinforced the perception that refusals to licence are not a pervasive problem.^[57]

18.48 One reason for the refusals cited in the Nicol–Nielsen Study was that exclusive licences had been granted to other companies. Competition between the patent holder and the company seeking to licence was also cited. In some cases, refusals also occurred because of an inability to agree on reasonable licence terms.^[58] Nicol and Nielsen commented that:

One interpretation of this data is probably that refusals to license were not encountered because often it did not get to the stage that licences were requested. This was acknowledged by many of our respondents. As reported elsewhere, researchers and companies stated that they avoided particular areas of research if patents were held by competitors, or if it looked as though obtaining a licence might prove to be too problematic … in line with the survey results a few interview respondents expressed frustration at difficulties in licensing-in enabling technologies, but these were greatly outnumbered by the number of respondents who had not experienced any problems. Some respondents complained that owners of research tool patents, while willing to license, unreasonably demanded reach-through royalties.^[59]

18.49 The need to licence-in patented technology may be a barrier to commercialisation if licences are not widely available. In particular, exclusive licences have the potential to be anti-competitive because they restrict access to important genetic materials or research tools.^[60] Compulsory licensing, discussed in Chapter 27, may provide some solutions to problems resulting from refusals to license. That chapter also recommends that an additional ground for obtaining a compulsory licence based on a competition test be included in the Patents Act (see Recommendation 27‑1).

Lack of experience with commercialisation

18.50 It has been suggested that there is a lack of appropriate commercialisation experience related to biotechnology within the Australian industry.^[61] A related problem is that Australia lacks a large enough pool of people with the skills to manage intellectual property effectively in the biotechnology context.^[62] This concern was raised in the Australian Science and Innovation Mapping Taskforce report, Mapping Australian Science and Innovation, whichstated that scientists, when taking on the role of Chief Executive Officer, ‘often do not have the specialist business skills to enable the company to survive in early-stage commercialisation’.^[63]

18.51 In consultations, Benitec Ltd noted that Australia’s ability to commercialise its research will improve as the country’s skill base grows as a result of experience.^[64] AusBiotech Ltd noted that there is a need to educate the research sector and industry further about intellectual property protection to ensure that future marketing applications for product development are considered early in the patent strategy.^[65] Intellectual property protection and technology transfer from research organisation are discussed in Chapter 17.

Lack of investment and venture capital

18.52 One possible barrier to effective commercialisation of genetic research within the biotechnology industry in Australia is the lack of long term venture capital funding.^[66] It has been suggested that it is very difficult to attract venture capital in Australia due to the lack of a mature venture capital base in this country.^[67] It has also been suggested that there is a need for venture capital with at least a five-year term.^[68]

18.53 Seed funding^[69] rates are also much lower in Australia compared with the United States. In Australia, the usual level of seed funding is around $1,000, while in the United States the level is closer to $1 million.^[70] In consultations, UniQuest suggested that higher funding at this stage enables United States companies to establish effective management structures, initially staffed by professional managers.^[71] Lack of investment can lead small, early stage companies to license their intellectual property too early in an effort to maintain cash flow. The original intellectual property may therefore be undervalued.^[72]

18.54 Insufficient early stage funding prevents new companies from establishing effective management structures.^[73] As a result, some biotechnology start-up companies are managed by academic researchers, rather than professional managers with experience in commercial negotiations and intellectual property management. As discussed in Chapter 14, most academic researchers do not possess the specialised skills necessary to manage a commercial venture. Consequently, some companies may be managed inexpertly and intellectual property exploited ineffectively.

18.55 Many Australian biotechnology inventions fail to be exploited effectively because of a lack of funding at the proof-of-concept stage. At this stage, an invention has been created and its commercial potential must be demonstrated to attract investment for its development into a marketable product. Passing the proof-of-concept stage involves demonstrating the commercial potential of an invention to attract investment for development of a marketable product.

18.56 In some cases, an invention at this stage will not be exploited at all. In others, the invention is licensed to an international company prematurely and its potential value to Australia is lost.^[74] Biotechnology Australia has suggested that this commercialisation gap ‘is widely recognised as the most critical barrier to biotechnology development in Australia’.^[75] For the most part, this issue is beyond the scope of this Inquiry.