Much of patent prosecution and opposition at the EPO is an advanced game of spot the difference. Subtle differences between the application as filed and the claimed subject matter can be fatal to a granted European patent. A recent decision from the Board of Appeal (T 1303/18) confirmed the high bar of a "clear and unambiguous disclosure" of the claimed subject matter in the priority document is necessary for a valid priority claim. Patent applicants should be aware that even the most subtle of differences between the claims and the priority document can be fatal to priority claim and thus to the validity of a patent in the case of intervening prior art.
Legal background: Priority for "the same invention" test
A European patent application may claim priority from a previous patent application for a period of 12 months from the filing date of the previous application (Article 87 EPC). When priority is claimed, the filing date of the previous patent application becomes the effective filing date (or priority date) of the previous application. Crucially, in order for a priority claim to be valid, the subject matter for which priority is claimed must be disclosed in both the earlier and later applications (Article 88 EPC, G 2/98).
Spot the difference |
The test for establishing whether both applications disclose the same invention is whether a skilled person can directly and unambiguously, using common general knowledge, derive the subject matter for which priority is claimed from the previous application as a whole (Case Law of the Boards of Appeal, II-D, 3.1.1).
Spotting the difference
In T 1303/18, the patent (EP 2215072) related to a polymorphic form of Rotigotine (Neurpo), a dopamine agonist approved for the treatment of Parkinson's disease and restless legs syndrome. The patent was granted with a priority date of 28 Nov 2007. The opponent argued that claim 1 as granted was not entitled to this priority date. The question before the Boards of Appeal was whether the priority application disclosed the subject-matter of the claimed invention. The Board of Appeal noted that this test should be satisfied to the same standard as required for assessing whether amendments to an application add matter (Article 123(2) EPC).
Claim 1 as granted claimed the polymorphic form of rotigotine as defined by a list of various functional properties, e.g. by its melting point and X-ray powder diffraction spectrum. However, there were some subtle differences between the definitions specified in the granted claim and that of the priority application. One difference, for example, was that the X-ray powder diffraction spectrum peak angles in the claim were defined as having an error margin of ±2, while the priority application specified peak angles with an error margin of ±0.1. For the Board of Appeal, this meant that according to granted claim 1, the peaks could be shifted so that they no longer matched the diffraction angles specified in the priority application. The Board of Appeal thus concluded that the claim was broader than the disclosure of the priority application, and that the polymorph specified in the claim was not clearly and unambiguously disclosed in the priority application.
The patentee argued that it would be clear to a skilled person that only two polymorphs of rotigotine were known at the priority date. As such, rejecting the validity of the priority claim amounted to insinuating that a third polymorph or rotigotine existed, i.e. a polymorph falling under the claim and not disclosed in the priority application. The patentee argued that a skilled person would clearly see this as being incorrect. The Board of Appeal rejected this argument, agreeing with the opponent that if the argument were to be accepted, the earliest priority would become retroactively invalid if a third polymorph was later discovered, thus running counter to the principle of legal certainty.
The patentee also argued that the opponent bore the burden of proving that the granted claims defined a polymorph different to the one disclosed in the priority document. The Board of Appeal resoundingly rejected this argument. For the Board of Appeal, it was the patentee that had introduced the difference between the application as filed and the priority document. As such:
"If, despite these differences, the appellant asserts that the compound of granted claim 1 is the same as that disclosed in [the priority application], it is the appellant who bears the burden of proving this. In fact, it is for an applicant to ensure, where priority from a previous application is claimed, that the same invention is defined, according to the required standard, in the subsequent application as filed" (r.2.13, see also Headnote)
The Board of Appeal therefore found the priority claim invalid. The claimed subject matter was thus found to lack novelty in view of prior use of the claimed polymorph that occurred between the priority date and the patent's filing date.
Final thoughts
The Board of Appeal in this case confirmed that the strict added matter provisions of the EPO apply not only to the application as files, but also for differences between the priority application and the subsequent application claiming priority. In a jurisdiction where the mere misplacing of a comma can be considered to add matter (IPKat), updates to your PCT or European application are made at an applicant's peril.
It is a pity that the Board did not accept to discuss the late-filed arguments relying on G1/15 (reason 5.7 of decision). It seems to me that the "partial priority" is crucial here. OK, these arguments are late-filed, but in my opinion, the patent benefits from the partial priority with the initialy disclosed ranges, so that the prior use would not be prior art...
ReplyDeleteThe boards are not very lenient with respect of late-filed submissions. The question of partial priority might have changed the outcome for the proprietor, but this argument should have been brought much earlier.
DeleteEven before the RPBA20, the mere fact that OP would have to be postponed was enough for late submissions not to be admitted.
I wrote the first comment. I fully agree that non-admissibility of late arguments is now common practice,but here, the outcome would have been the opposite!!! The patent is revoked because of provisions directed to the economy of procedure... Hard to hear...
DeleteIt is hard to see how G1/15 and partial priority could have saved the patent because the prior art was a public prior use in the priority year. Even had the claim, defining peaks to within +/-2, been assigned partial priority to the embodiment defining peaks to within +/-0.1, the public prior use would have remained relevant for the inventive step of the parts of the claims defining the peaks to within +/-2 but not within +/-0.1. No inventive step could be acknowledged for this.
DeleteThe claim would have needed to be limited to the narrower tolerances disclosed in the priority document.
Hard to hear, but hard reality!
DeleteDo not forget that the RPBA20 were drafted with the aim to increase the throughput of the boards. But even under the RPBA07 it would have been too late!
it is more than doubtful that G 1/15 could have saved the patent.
DeleteThe board has shown that two selections were needed to go from the priority document to the patent. In such a situation, identity of invention is simply not given.
Reducing the claim to the tolerances given in the priority amendment, would not have helped much.
In claim 1 as granted, the claimed compound is said to be a polymorphic form II of rotigotine and is defined by at least one of four properties.
ReplyDeletein particular property (a) a certain X-ray powder diffraction spectrum comprising four peaks with an error margin of ± 0.2, said to be measured with Cu-Kalpha irradiation (1.54060 Å).
For the board and in view of the established differences, it followed that, in order to arrive at the four peaks stated in claim 1 as granted, at least two selections within D49 are needed:
- a first selection for defining the compound of D49 by means of a number of four peaks within the X-ray powder diffraction spectrum; and
- a second selection for arriving at the specific diffraction angles (°2theta) mentioned in claim 1 as granted for these four peaks.
D49=priority document did not contain any pointer towards these two specific selections. The priority was thus not valid.
The board did thus not simply base the lack of identity of the invention on the different error margins of ±2 in the patent and ±0.1 in the priority application.
The board held that the error margin has been broadened as compared with the error margin of ±0.1 disclosed in the priority document. This means that, according to claim 1 as granted, the peaks can be shifted so that they no longer match the diffraction angles (°2theta) disclosed in D49. No disclosure of this broadening of the error margin can be found in D49.
Identity of the invention was not given. If the proprietor claims identity of invention, there is nothing abnormal in shifting the onus of proof to the proprietor.