An EPO Board of Appeal left the pharmaceutical sector much to ponder in February this year when it confirmed an earlier EPO Opposition Division’s decision to revoke Bristol-Myers Squibb’s European patent EP1169038 on the ground of a lack of inventive step (T488/16).

This has raised eyebrows for generics and big pharma alike, as many would have predicted that the subject of the patent, dasatinib (Sprycel®), would be safe from revocation, protected as it was by a single narrow claim to the commercially successful cancer drug.

The background

Like many new chemical entity cases, dasatinib (a protein tyrosine kinase inhibitor worth $1.6 billion per year) was one of many compounds disclosed in the application as filed, which (critically, as it turned out) did not contain any activity data for a single one of the 580 compounds disclosed.

The novelty of the compound was not disputed, and the whole case hinged around plausibility for inventive step, specifically to what extent an applicant/patentee can rely on activity evidenced only after filing the original application.

The case law

Current EPO case law mandates that as long as a technical effect is at least made plausible by the disclosures of the application as filed, then the applicant/patentee can use post-filed data to corroborate the technical effect. The current benchmark for plausibility was set by T1329/04 (Factor-9/JOHN HOPKINS), which rejected the possibility of using post-published as the sole evidence for a technical effect and concluding that the disclosure of the application itself must at least make it plausible that the invention indeed solves the problem it purports to solve.

In a fast-paced, cut-throat industry such as pharmaceuticals, patent filing practice can include filing applications to numerous compounds before a lead compound is chosen. Reliance may be placed on post-filed data to convince patent office examiners of an inventive step.

The outcome

In this case (T488/16) the Board of Appeal decided that in the complete absence of any activity data, the burden of proof fell upon the patentee to show the technical effect was plausible at the time of filing.

The Board found a general statement in the application that each compound had been found to show activity was not sufficient to render a technical effect plausible, and thus refused to admit any post-published data in the evaluation of inventive step. Using the typical EPO "problem-solution" approach, the Board found that the problem to be solved was merely the provision of a further chemical compound, which was not considered inventive. The patent was therefore revoked for lack of inventive step.

Our insights (and some practical notes)

Interestingly, the Board was not swayed by the patentee’s argument that the EPC does not require experimental proof and instead insisted that experimental data is needed in cases where the technical effect relied on is “neither self-evident nor predictable or based on a conclusive theoretical concept”. This statement suggests that the EPO may be shifting towards favouring quantified technical effects and highlights the wisdom of filing when an invention is supported by enough data to be at least made plausible.

This Decision also suggests that the EPO is taking a tough stance on those who are on the wrong side of the narrow line between patent protection and a trade secret. Attempts to preserve confidentiality by withholding important technical information on filing may well backfire on those looking to gain an advantage over their competitors. For those applicants that do not wish to reveal too many details of their lead compound, it is sensible to at least define a threshold of activity in a given assay, as plausibility attacks could soon become the vogue at the EPO.

You can read the full Decision here.