ALEX WILSON

Senior Associate, Bioinformatics Group, Bristows

The Bioinformatics Sector

Vast sums of money have been invested in research in the bioinformatics field, from the sequencing of the human genome and proteome, the creation of databases incorporating the resulting sequence data, to the development of software and hardware used to make sense of that data. Yet the intellectual property protection available in return for those investments has, until recently, seemed elusive.

The bioinformatics sector first came to the general public’s attention as a result of the human genome project. In reality, bioinformatics, that is, the use of computational tools and databases in relation to biological, medical and health data, had been a discipline in all but name for over a decade. Public and private databases, such as BLAST and Genbank, which allowed sequence comparisons on a substantially less grandiose scale, were available well before the human genome project. In addition to the publicly funded human genome project, companies such as Celera, Human Genome Sciences and Incyte have devoted very significant resources to the sequencing of human and other species’ genomes. Although their business models have begun to diverge, those companies started off as essentially database providers intending to recover their investment by charging subscription fees for access to their data. As the volume of data produced grew exponentially, companies specialising in mining the data, such as Lion Biosciences, emerged, developing algorithms and software for making sense of the tsunami of data which now also include proteomic, clinical and related data. Most recently, the availability of a large number of disparate data sources and compatibility issues has bred a number of so-called bioportal companies such as Bio Wisdom.

Perhaps as a result of their proximity to the patent-conscious biotechnology community, bioinformatics companies are increasingly looking towards patent protection in an attempt to capitalise on their investments. In addition to patent protection for gene-related inventions and drug-screening methods, other rights such as database rights, confidential information and copyright may be available to database companies. This article explores the current practice of the European Patent Office (EPO) and its US counterpart (USPTO) in relation to patents in this area.

The so-called genomics companies referred to above have been some of the most prolific patent applicants in the past few years. For example, Human Genome Sciences has filed tens of thousands of patent applications and so far has been granted over 196 patents relating to human-based gene inventions. At one stage Incyte reported that it had filed patent applications covering ‘portions of more than 50,000 genes’ including patent applications on 7,000 full-length genes. Although these applications often relate to the sequence data identified by the use of bioinformatic tools, the intention of these companies is primarily to obtain patent coverage for applications of that sequence data, typically for therapeutic purposes.

In addition to the patents for full-length genes, many patent applications have been made in the United States which relate to partial sequences known as expressed sequence tags (ESTs), that is, the fragment of the genome which is actually expressed by the organism rather than those parts which effectively lie dormant. ESTs can be generated using automated sequencing technologies which have been around for over 15 years. In fact, the US National Institutes of Health (NIH) and the UK Medical Research Council (MRC) filed the first EST applications in the early 1990s. As a result of the public outcry at the time, the applications were withdrawn and therefore the patentability of ESTs did not have to be considered by the USPTO until the late 1990s when the first EST patents were granted.

The principal objection to the patentability of ESTs in the United States has been that they lack ‘utility’. Utility can be described as the actual or potential uses of the invention. The first attempt to standardise USPTO examination in relation to biotechnology inventions was made in 1995 with the introduction of Utility Guidelines which required all patent applications to describe at least one specific utility which was credible. Examiners generally viewed this as meaning that it would not be sufficient for a statement to be included in the application that it had ‘biological activity’. The utility which was asserted had to be possible in the ‘real world’. For example, the utility could not contradict known natural laws or be ‘throwaway’, such as using a polypeptide as an amino acid source, or a molecular weight standard. However, the 1995 guidelines did not prevent significant numbers of EST-based patents being granted. The burgeoning numbers of such applications led to further guidelines being issued by the USPTO in early 2001. Under these new guidelines, utility now needs to be substantia as well as credible and specific. This requirement is apparently directed at excluding throwaway utility but, at least initially, the term ‘substantial’ raised as many questions as it answered. It appears that the effect of the guidelines has been that genome-based patent applications which disclose only theoretical utilities are now being rejected. Given that the specific utility of a particular gene or polypeptide can rarely be demonstrated until there has been a sufficient characterisation of the function of the gene or its expression product, the new guidelines are likely to slow the rate at which patent applications will be filed. Critics of the new utility threshold argue that it comes too close to a requirement that the stated utility is commercially viable at the time the application is filed. Clearly, in emerging technologies such as genomics and proteomics, such a requirement would be quite onerous, particularly where there is pressure to put the sequence data into the public domain as soon as possible. However, the guidelines make it clear that the benefit of the doubt will be given to the applicant. If an applicant asserts a substantial specific utility for a gene sequence this will only be rejected if the examiner believes that there are technical grounds for doubting the assertion.

The uncertainties regarding the European Biotech Directive1 seemed finally to come to an end in October 2001 when the European Court of Justice rejected the Dutch Government’s challenge to the Directive.2 There should now be no further obstacle to its incorporation into national laws. EU Member States should have incorporated the Biotech Directive into national laws by 30 July 2000. However, despite robust encouragement from the Commission, to date only Denmark, Finland, Greece, Ireland and the United Kingdom have done so. France and Germany are the most obvious absentees from this list and are still debating whether to implement the Directive. In the case of France, draft bioethics legislation is being considered which would prevent the patenting of gene sequence inventions, in direct contravention of the Directive. The European Patent Office was much more rapid than most EU Member States in embracing the Biotech Directive, despite the fact that as a non-EU organisation4 the EPO is not bound by the Directive. The EPO has introduced amendments to the European Patent Convention (EPC) which broadly mirror those of the Directive in relation to the patentability of biotechnological inventions and the Directive is now used as a supplementary means of interpretation by the EPO.

Under the Directive, simple discoveries of one of the elements of the human body, including the sequence or partial sequence of a gene, are not patentable. Conversely, an element isolated from the human body or otherwise produced by means of a technical process can constitute a patentable invention. Such an isolated element could be the sequence or partial sequence of a gene or protein, even if that sequence is identical to the natural element, provided that the industrial application of the sequence or partial sequence is disclosed in the patent application. The Directive excludes from patentability those inventions which would be contrary to morality, a basis of attack which has been the focus of many unsuccessful challenges to the validity of biotech patents.

The nearest equivalent to a utility requirement under the EPC is the requirement for ‘industrial application’. It is fair to say that lack of industrial application had been an unpopular ground of invalidity in the EPO until the amendments were made to the EPC to incorporate the Directive. Recently, the EPO has shown a tendency to fashion the industrial application requirement into a quasi utility requirement. The EPO’s argument is based on the fact that Rule 23e(3) introduced a requirement that the industrial application of a sequence or partial sequence of a gene must be disclosed in the patent application. This requirement was clearly directed to the prevention of the patenting of bare sequences in circumstances where the applicant had not established the function or use of the gene in question. In the past, the need to disclose an industrial application would not have caused many applicants difficulty, as industrial application under Article 57 EPC imposes a low threshold, requiring only that the invention can be made or used in any kind of industry. Any gene sequence could be made in industry and therefore such applications would previously have satisfied Article 57. However, the Opposition Division in the recent Icos decision held that it was not sufficient simply to list potential uses of the invention in the application. The uses disclosed in the application must not be speculative, they must be specific, substantial and credible. Interestingly, those words used by the Opposition Division have no basis in either the EPC or the Implementing Regulations, but they are identical to those used in the USPTO examination guidelines discussed above. This convergence of opinion may be a result of the Trilateral Project between the USPTO, the EPO and the Japanese Patent Office. It is perhaps surprising that in the absence of appropriate amendments to the EPC, the EPO is apparently imposing a higher threshold for industrial application in relation to biotechnological inventions than for other inventions. Such a criticism could not be levelled at the USPTO Guidelines which apply to all fields of technology. A further problem with the EPO’s approach is that it may be inconsistent with the Directive and/or TRIPs. In particular, the Directive states that it is without prejudice to the Member States’ obligation pursuant to TRIPs. TRIPs requires that ‘patents shall be available and patent rights enjoyable without discrimination as to … the field of technology …’. The EPO approach could be seen as discriminating against biotechnology inventions in breach of this provision of TRIPs.

Until the Enlarged Board of Appeal has had an opportunity to consider the new industrial application provisions of the EPC, applicants for gene sequence-based patents will be well advised to specify substantial, credible and specific uses for the invention in the application.

It is often said that the sequencing of the human genome was an achievement in computing rather than in molecular biology. Since the development of that enabling technology, a number of bioinformatics companies have been active in creating new algorithms and software for accessing and using the available data, for example for producing drug targets. Those software companies have been seeking to obtain patent protection for the fruits of their research.

While the USPTO was being inundated with applications for EST patents, it was also receiving a wave of applications relating to business methods and software. The USPTO and the US courts have been relatively accepting of these applications. On the other hand, the EPO has been seen as considerably less enthusiastic. In particular, Article 52(2) EPC excludes from patentability ‘… methods of doing business … and programs for computers …’. However, that restriction only applies to the extent that the application relates to programs for computers as such. In other words, a patent claiming the software itself would not be patentable, but if the invention makes a non-obvious ‘technical contribution’, it may be patentable even if the subject-matter defines or involves a computer program. The concept of technical contribution is relatively opaque, but it seems that a computer program would have a technical character if, when run, it caused a technical effect which goes beyond the normal physical interactions between program and computer. In the bioinformatics field, a technical contribution could lie in a particularly efficient method of clustering proteins or of identifying a drug target. EPO case law in this area is generally becoming more accepting of so called software patents but is still some way behind the United States in this respect. The European Commission, and the UK Patent Office have held recent public consultations with industry and other interested parties. Following those consultations the European Commission has recently proposed a Software Patenting Directive, with the aim of harmonising the currently divergent approaches of the courts of the Member States to such patents. The draft Directive has been seen by many as a disappointment as it continues with the EPO’s status quo rather than adopting a more liberal approach along US lines.

The bioinformatics industry itself is expected to grow to US$1.7 billion by 2007. However, the most valuable aspects of this industry lie in the therapeutic products which will be identified with the use of bioinformatics tools. For this reason, companies active in the bioinformatics field have tried to establish mechanisms for reaching through to the bioinformatics-derived therapeutic products to obtain royalties based on the sales of those products rather than on the bioinformatics tool used for their identification. Some of the database providers such as Incyte initially attempted to achieve this by contractual means. One form of their licence agreements with academic institutions included a provision that ownership of any intellectual property in products made using their data would vest in the database provider. Understandably, this type of provision proved to be unpopular and has not been adopted as an industry standard.

Patent attorneys have also attempted to achieve this reach-through in their claim drafting. However, as the patents often relate to screening methods which are some way upstream from the therapeutic product, it is easy to imagine that in many cases the claims will suffer from insufficiency problems. On the other hand, traditional claims directed to the bioinformatics method of screening can potentially be circumvented by carrying out the method outside a jurisdiction in which the patent is in force. Even if the acts are carried out in a jurisdiction in which the screening patent is in force, the damages recovered by the patentee may be disappointing. This is because there remains some doubt in a number of jurisdictions whether, if such a research tool patent is infringed, the proprietor would be entitled to recover damages based on the ultimate therapeutic product rather than on the screening method itself.

There is of course no such thing as a typical bioinformatics patent as the field is extremely broad. Companies such as Human Genome Sciences are increasingly focusing on the therapeutic uses of gene sequence-derived inventions, while other companies such as Incyte and Oxford Glycosciences are filing applications in relation to the screening methods.

As might be expected, the USPTO has generally reacted favourably to applications for screening methods which in many ways resemble the business method patents which were granted to ‘dot.com’ companies. The EPO has again been slow to react to this new field with, to the author’s knowledge, no such applications having been granted to date. There is evidence to suggest that even at this early stage, applicants who have filed these types of applications in the United States are failing to apply for equivalents for such patents in the EPO, perhaps because an unfavourable outcome is expected. That said, such an approach may be unduly pessimistic as inventions which provide a technical contribution should be entitled to patent protection in the EPO. For example, the invention described in the claim in the box below might be found to provide a technical contribution by providing a search template for a lead molecule, but query whether the patentee should be entitled to reach through to the product of that method.

The practices of the EPO and USPTO are converging in relation to gene sequence patents. The EPO is focusing on the requirement for industrial application while the USPTO’s approach is centred on utility. However, in effect the same criteria are being applied: substantial credible and specific use must be shown for the invention.

The Commission’s proposal for a Software Patenting Directive represents a further step away from the current more liberal USPTO approach. Bioinformatics companies should not despair, as many of the complex software screening inventions may have the requisite level of technical contribution to be entitled to patent protection under the proposed Directive.

A method of generating a molecular interaction search template for a lead molecule predicted to be capable of interaction with a target protein, said method comprising the steps of:

• Predicting the configuration of a binding site in said target protein;

• Dividing the binding site into a plurality of grid points;

• Generating a three-dimensional density map of preferred atom-atom contact distributions in the binding site; and

• Generating the molecule interaction search template from said three-dimensional density map.

To make any comments on this article, or to ask a question of the author, please contact the publisher. If you would like to submit an article please subscribe to our PL Intelligence service.

The opinions expressed in the articles published in this section do not necessarily reflect those of Pharmalicensing or UTEK Corporation. No actions including proposals to or agreements with other companies should be taken by any reader without obtaining specific business or legal advice. Neither the publisher nor the authors accept any liability for any actions or activities undertaken by any reader or other third party as a consequence of these articles or for any errors or omissions therein.