The EPO’s Board of Appeal recently revoked a patent covering Bristol-Myers Squibb’s blockbuster oncology drug dasatinib (marketed as Sprycel®). This decision (T488/16) provides valuable guidance into avoiding the pitfalls, and exploiting third party vulnerabilities, associated with the issue of plausibility. Plausibility is central to a number of statutory requirements to patentability at the EPO including inventive step, sufficiency and industrial applicability. For example, if the technical effect of an invention is not rendered plausible by the patent application, the technical effect may not be considered in the assessment of inventive step. Further, post-filed data may only be used to support a technical effect which was made plausible in the application as filed.
The present decision relates to Bristol-Myers Squibb’s now revoked patent EP1169038. The granted patent comprised a single independent claim defined by a Markush formula which covered an extremely wide range of compounds allegedly useful as protein tyrosine kinase (PTK) inhibitors. The Opposition Division had concluded that, at the filing date of the application, the claimed compounds had not been plausibly demonstrated to be PTK inhibitors suitable for the treatment of cancer, which was the technical effect the patentee relied upon for inventive step. The Opposition Division therefore ruled that post-published evidence could not be used to support inventive step and the patent was deemed to lack an inventive step.
In appealing the decision of the Opposition Division, the patentee amended claim 1 to cover a single chemical compound (dasatinib). The patentee argued that the EPC does not require experimental proof of a technical effect and that a statement of alleged activity provided in the application as filed was enough to meet the plausibility threshold. Whilst the Board of Appeal agreed that it is not always necessary to include experimental data in an application, it is however essential that it is shown that the technical problem underlying the invention was at least plausibly solved at the filing date. The Board judged that it is not acceptable to draw up a generic formula covering millions of compounds, vaguely indicate an “activity” and leave it to the skilled reader to establish which compounds may be suitable to treat a certain disease.
In the Board’s decision, it is emphasised that the issue is not a lack of in vivo or clinical data, but rather the absence of any verifiable data with regard to the alleged technical effect. The Board therefore upheld the decision of the Opposition Division revoking the patent. It is unclear from the present decision whether the plausibility threshold would have been altered if the application had claimed a smaller number of more structurally related compounds. However, it is noteworthy that limiting the claims to a single compound during the appeal proceedings was not considered to lower the plausibility threshold.
The present decision illustrates the ongoing importance of striking the correct balance between obtaining the earliest priority date for an application and having enough experimental data to make the technical effect relied upon plausible. This decision confirms that, for at least pharmaceutical patent applications, it may be beneficial to delay filing until at least some evidence of the alleged technical effect is available, preferably across a range of representative compounds.
The present decision also indicates that the sheer number of compounds included within a Markush formula may raise the plausibility threshold. Therefore, if protection is sought for a large group of unrelated compounds that can be defined by a single generic formula, it may be worthwhile pursuing separate applications covering certain subsets of compounds instead of a single application covering all possibilities, since limiting a patent to a specific compound at a later stage does not appear to lower the plausibility threshold. This may be particularly important where post-published evidence demonstrates that certain compounds covered by the generic formula do not demonstrate the technical effect relied upon.
The present decision confirms that plausibility remains an important issue at the EPO. However, it remains to be seen whether the threshold for plausibility has actually been raised by the present decision and whether a similar threshold would be applied to a patent covering a smaller number of compounds.