UK: Deciding When To File A Patent Application In Europe

Originally published in Nature Biotechnology , Vol. 25 No. 6, June 2007.

Guidelines for would-be patent applicants in Europe's 'first-to-file' system

Deciding on the right time to file a patent application can be a difficult decision in any technical field, but it is particularly difficult in competitive and rapidly developing areas of biotech. Would-be patentees must make a decision whether to delay filing until a substantial amount of data have been accumulated to support the invention and risk being 'scooped' by a competitor, or alternatively file early to secure a filing date ahead of the competition and risk the patent application being rejected as lacking adequate experimental support. Choosing the right time to file is particularly difficult in jurisdictions such as Europe, which operate a 'first-to-file' rather than a 'first-to-invent' system.

Patent applicants have tried to circumvent this problem by filing early on the basis of limited experimental data to establish a filing date ahead of potential competitors, on the understanding that it may be possible to submit further experimental data at a later date to show that the invention did in fact meet all the requirements for patentability as of the filing date.

Various recent decisions of the European Patent Office (EPO; Munich) have cast doubt on the ability to rely on late-filed experimental data (that is, data submitted to the EPO after the filing date of the patent application) to support patentability. This article summarizes the effects of these decisions, and identifies the circumstances in which applicants should not expect to rely on late-filed experimental evidence to support patentability before the EPO.

Sufficient disclosure requirement

Biotech inventions must satisfy the basic patentability requirements common to all technical fields: that is, the invention must be novel and nonobvious in comparison to the prior art. The invention must also be supported by sufficient technical disclosure to enable one skilled in the art to practice the invention over the full scope claimed without undue burden.

It is well established in the case law of the EPO Boards of Appeal that in order to meet the requirement for sufficient disclosure (Art.83 EPC) the application as filed must disclose at least one way of carrying out the invention. If the claims are particularly broad, then it may be necessary to provide more than one example to ensure that the technical disclosure of the application is sufficient to enable a skilled reader to carry out the claimed invention over the whole range claimed.

If the technical content of the application is judged fundamentally insufficient as of the filing date, it is not possible to remedy the situation by submitting late-filed experimental evidence. This principle is illustrated in an EPO Technical Board of Appeal decision concerning a claim to use of a peptide having insulinotropic activity for the treatment of diabetes mellitus1. In this case, the peptide was defined as consisting of a 31-amino acid sequence (GLP1(7-37)) but lacking one amino acid. The application as filed lacked any data to show that any of the claimed peptides were effective in the treatment of diabetes, but described an assay that could be used to test whether particular peptides would be effective. The applicant maintained that it was well within the capabilities of the skilled reader to make any one of the 31 peptides falling within the claim and test it for the required activity using the disclosed assay. Several postpublished documents had confirmed that at least one of the claimed peptides (GLP-1(7-36)) had the claimed biological activity.

The Board came to the conclusion that the technical disclosure in the application as filed was fundamentally insufficient, because it placed an undue burden on the skilled reader to make and test all of the 31 peptides falling within the claim to find a peptide with the required activity. The postfiled evidence confirming that at least one of the claimed peptides did, in fact, have the required activity could not remedy the lack of sufficiency present at the filing date.

The requirement for sufficient disclosure at the filing date does not mean that it is always necessary to include actual experimental results to demonstrate that the invention works as claimed. In certain situations, it may be sufficient to include a hypothetical example, provided that the skilled reader can follow this example and arrive at the claimed invention without undue burden or exercise of inventive skill. Nevertheless, applicants should be aware that use of 'paper examples' alone can lead to a reversal of the burden of proof in establishing that the disclosure of a patent is sufficient, specifically in European Opposition proceedings, as illustrated by a Board of Appeal decision2 concerning the display of proteinaceous binding domains on the surface of filamentous phage. In this case, the application as filed contained only a hypothetical experimental example and did not include any experimental data to confirm that the invention would work. Postfiling data submitted by the patentee was based on an experimental protocol that deviated from the hypothetical protocol in the application. The Appeal Board confirmed that, although use of hypothetical examples is not prohibited, if the only example is explicitly described as a hypothetical example that has clearly not been carried out, the burden of proof is on the patentee to show that what is described actually works.

Potential applicants should therefore be aware that although it is possible to file on the basis of a hypothetical example only, and adduce technical evidence after the filing date to show that the example works, problems will arise if the hypothetical example does not in fact work as it is written, or if substantial variation from the protocol as written, amounting to 'undue burden' or 'exercise of inventive skill', is required to make it work. Moreover, it appears from later EPO decisions that there are some technical fields in which hypothetical examples can never be enough to satisfy the requirements of the European Patent Convention (EPC). There are some situations where experiments will absolutely be required to show that an invention has been made. Falling into this category are inventions relating to therapeutic treatment, claimed in Europe as new medical uses.

In one key case3, the EPO Technical Board of Appeal considered the patentability of a claim to the use of a steroid hormone or steroid hormone analogue, which fails to promote transcriptional activation of glucocorticoid receptor or retinoic acid receptor genes, for the preparation of a pharmaceutical for the treatment of AP-1 stimulated tumour formation, arthritis, asthma, allergies and rashes. The application had been refused by the Examining Division for a lack of sufficient disclosure in relation to this medical use claim. Although the application disclosed a screening method for finding steroid molecules with the required properties, no such molecule was identified or disclosed and no data of any kind were presented indicating that such a hormone could have an impact on any of the diseases is was claimed to treat.

The applicant provided postpublished evidence that steroid hormones having the mechanism of action required for the medical uses claimed had indeed been identified using screening methods identified in the patent application, and argued therefore that by carrying out the teaching in the parent application, one would necessarily obtain pharmaceutical compositions that would have the claimed medical utility. The Appeal Board did not agree with this approach, stating that sufficiency of disclosure must be satisfied at the effective date of the patent. In the case of a claim to a medical use of a pharmaceutical composition, attaining the therapeutic effect was a functional feature of the claim. Unless this was already known to the skilled person at the priority date, the application must disclose the suitability of the product for the claimed therapeutic application to satisfy the requirement for sufficient disclosure. This did not mean that human clinical trials, or even animal data, were necessarily required, but there must be some evidence that the claimed molecules actually have an effect on the metabolic mechanism involved in the disease. The presence of a cause-and-effect relationship must be made 'plausible'. Even showing a pharmaceutical effect in vitro may be sufficient, if for the skilled person this observed effect directly and unambiguously reflects a therapeutic application. Such rudimentary evidence may be backed up with further evidence produced later.

This case established that although it is not always necessary to have clinical trial data, or even animal data, to support a claim to a new medical use of a known substance, it is not sufficient merely to state that "substance X can be used to treat disease Y". Such statements must be backed up by some form of experimental data, whether animal data or data from an accepted in vitro model, which backs up the claim to medical utility. The supporting data must be present in the application as filed to provide acceptable disclosure of 'the invention' as claimed. Thus, although it may be possible to file on the basis of animal data, or even in vitro data, and back this up during prosecution with in vivo data if and when required, it is not possible to submit data generated after the filing date to back up a simple statement of medical utility.

The implication of the EPO case law is that even if the specification as filed provides 'sufficient' technical teaching to enable a skilled person to make and use "substance X" in the treatment of disease Y, and it could be later shown that by following this teaching the claimed treatment would indeed be successful, the requirements for patentability would not be met if the specification as filed lacked the necessary experimental evidence to convince a skilled reader that the treatment could reasonably be expected to work.

Showing the inventive step

Going beyond the requirements of sufficiency, a more recent Technical Board of Appeal decision4 has cast doubt on the ability of applicants to adduce post-filing date evidence in support of inventive step, even if the basic disclosure is sufficient. The case in question concerned a patent application in the name of Johns Hopkins University (Baltimore) with claims directed to polynucleotides encoding polypeptides having the activity of growth differentiation factor (GDF)-9, a putative further member of the transforming growth factor (TGF)-² superfamily of growth differentiation factors.

The applicant had isolated a novel polynucleotide molecule encoding the GDF-9 polypeptide. It was not contested that the polynucleotide and encoded polypeptide represented novel molecules. The applicant claimed that the GDF-9 molecule represented a further member of the TGF-² superfamily, although its sequence was only very weakly conserved, exhibiting no more than 34% homology with known members of the TGF-² family, and the protein contained only six cysteine residues instead of the seven cysteine residues that were a key characterizing feature shared by all members of the TGF-² family. The application as filed did not contain any functional data characterizing GDF-9 as a growth differentiation factor of the TGF-² family.

The EPO assesses inventive step in terms of a solution to a technical problem arising in the closest prior art. In the Johns Hopkins case, the closest prior art described the identification of a different novel member of the TGF-² superfamily, GDF-1. The problem allegedly solved by the invention was therefore defined as providing a new growth differentiation factor of the TGF-² family. The applicant maintained that inventive step should be acknowledged on the basis of the unexpected structural properties of GDF-9, namely the weak homology with known members of the TGF-² family, the fact that it contained only six cysteine residues instead of seven and the limited expression restricted to ovarian tissues. The applicant also filed numerous postpublished documents as further evidence of the role of GDF-9 as a growth differentiation factor. This evidence claimed to support the description in the application as filed of the presumed functions of the GDF-9.

The Board of Appeal considered that on the basis of the information presented in the application as filed, GDF-9 could not be clearly and unambiguously identified as a member of the TGF-² superfamily because of the low sequence homology with known members of the family and the presence of only six cysteine residues rather than the characteristic seven residues found in family members. The application as filed contained no experimental evidence to support the predicted function as a growth differentiation factor. The Board concluded that as structural features alone failed to allow GDF-9 to be identified as a member of the TGF-² family and the application as filed did not contain sufficient evidence to make it "at least plausible" that a solution had been found to the problem of identifying further members of the TGF-² family, then inventive step could not be recognized. The postpublished evidence establishing that GDF-9 was indeed a growth differentiation factor, as the applicant had claimed all along, could not be relied upon to establish inventive step. The Board distinguished the situation where supplementary postpublished evidence may in some circumstances be taken into consideration to support technical evidence given in the application as filed, from the situation where postpublished evidence provides the sole basis to establish that the application does indeed solve the problem it purports to solve.

The logical conclusion from the Johns Hopkins case is that applicants who are deemed to have "speculated" as to the function of a novel molecule but do not present technical evidence in the application as filed to support the claimed function may find their claims refused for lack of inventive step, even if it can be established by postfiled experimental evidence that the "speculative" function was correct all along. The problem with this approach is that it introduces a subjective test to determine whether it was "at least plausible" that the invention provided a solution to the problem that was purportedly solved based on the content of the application as filed.

Indications to date are that EPO Technical Boards of Appeal are following the decision issued in the Johns Hopkins case, based on a subjective assessment of the "quality" of the experimental evidence included in the application as filed. Several subsequent EPO Appeal Board decisions5-7 all refer to the Johns Hopkins case but are distinguished from it on the basis that the supporting data in the application as filed is of "better quality". The implications are that if the quality of supporting experimental data included at the filing date is judged inadequate to satisfy the "at least plausible" test, then experimental results submitted after the filing date are unlikely to be of assistance. This has been interpreted by many patent practitioners in the Biotech field as introducing a requirement for actual reduction to practice as of the filing date of a European patent application, a requirement that finds no explicit basis in the EPC.

Conclusions

The message to prospective applicants from the EPO case law is that supporting experimental evidence may be required (i) to enable others to perform the invention over the full scope claimed, and/or (ii) to show that an invention has been made, in comparison to the prior art.

It is not possible to submit late-filed data to remedy a fundamental lack of sufficiency in the application as filed. Nevertheless, the EPO case law does not preclude submission of experimental evidence generated after the filing date to show that a paper example works. It is also possible to submit experimental evidence generated after the filing date to show that the invention works across the whole range claimed, provided the specification as filed contains sufficient technical disclosure to satisfy the requirements of Art. 83 EPC (e.g., at least one way of performing the invention is adequately described in the application as filed). With certain inventions, notably inventions based on therapeutic treatment, the requirements for sufficient disclosure go beyond enabling skilled readers to "make and use" the invention. The application as filed must contain some evidence of therapeutic efficacy. Even if the requirements for sufficient disclosure are met, experimental evidence may still be required to show that an 'invention' has been made in comparison to the prior art. It is not possible to use post-filing date experiments to show that an 'invention' solves the technical problem it purports to solve in comparison with the prior art if this is not "at least plausible" based on the content of the application as filed.