In our first article in this series on Australian and New Zealand law, we reported on the current law on amendment allowability following Australia’s “Raising the Bar” reforms and the decision of the Federal Court of Australia in CSIRO v BASF Plant Science GmbH and how that decision has been applied by the Australian Patent Office.

In this second article, we highlight a troubling aspect of Beach J’s decision in CSIRO v BASF Plant Science GmbH, suggest how it might be used to invalidate a large number of patents granted under the pre-Raising the Bar legislation, and give one example of where it has already gained traction.

For patents/applications for which examination was requested before 15 April 2013 the CSIRO v BASF Plant Science GmbH¹ decision has reagitated an issue thought to be resolved and has suggested an additional set of criteria for a sub-set of cases. As explained below, it would be prudent for patentees of valuable older patents which were amended during prosecution to review the strengths and weaknesses of the granted claims in light of this decision.

In spite of the earlier reference to the Explanatory Memorandum,² which clearly showed that the legislators thought that there was a difference between the old and new versions of the Patents Act 1990, there was one paragraph in Beach J’s decision which sent shivers down the spines of these authors.

While admittedly in obiter, at [221] Beach J posed:

Are the requirements of s 102(1) now stricter?

If there is a difference in scope, then the distinction may be subtle.

Essentially, Beach J was suggesting that the previous requirement for “in substance disclosure” was very similar to the current requirement that, as a result of the amendment, the specification does not “claim or disclose matter that extends beyond” the disclosure of the original specification.

Beach J supported the suggestion on three bases:

1. By reference³ to the decision of Pumfrey J in Palmaz’s European Patents (UK)⁴ (Palmaz), in which Pumfrey J used the expression “in substance disclosed” in a part of the decision discussing “intermediate generalizations”;
2. By reference⁵ to the 13th edition (published in 1982) of Terrell on The Law of Patents (Terrell), in which the authors noted that “the then new section 76(2) would likely be construed “more or less the same as the requirement under the old law that there must be a fair basis for the amendment in the unamended specification””; and
3. By reference⁶ to the decision of the Full Federal Court in Les Laboratoires Servier v Apotex⁷ in which Emmett J tested whether an amended claim feature “links back” to the original disclosure of that feature in isolation.

Respectfully, these authors believe that Beach J’s three bases of support for the suggestion that the current requirements of s 102(1) are not any stricter than the previous requirements are misplaced for the following reasons:

Palmaz

The part of Palmaz referred to by Beach J was not considered, let alone upheld on appeal – in fact Aldous LJ said that it was not necessary to decide that issue at all and he declined to do so (Henry LJ in agreement). Still further, decisions subsequent to Palmaz that were also referred to by Beach J did not advance any closer relationship between the expression “in substance disclosed” and “intermediate generalization”, and in fact applied the law less narrowly than Pumfrey J had, determining that the specification does not need to specify that the newly amended combination of features provides any inventive significance before it could be allowed, which was contrary to the position that Pumfrey J had put forward.

Further, in the paragraph immediately preceding that expression in Palmaz, Pumfrey J explains that an intermediate generalization is one where features are isolated from a particular context in which they are originally described. Pumfrey J then describes that “feature H” is “in substance disclosed” since it had not been taken out of context. When describing the amendment to incorporate “feature I”, Pumfrey J stated that this was “in a different category”, had been isolated from the context in which it was originally described, and was therefore an “intermediate generalization” which was unallowable.

As such, we prefer an alternate reading of Palmaz – namely that the expression “in substance disclosed” has no relationship with “intermediate generalization”, that the two were not equated at all, and that Pumfrey J used the expression plainly, rather than to deliberately link the criteria for the UK Patents Act 1977 to the UK Patents Act 1949.

Terrell

Terrell in fact suggested that fair basis was an appropriate test for determining whether an intermediate generalization was allowable, not some stricter assessment. On that basis Terrell stood for the proposition that the Australian High Court’s liberal approach in Lockwood v Doric⁸ was the appropriate test.

It is also worthwhile noting that the version of Terrell relied on by Beach J was authored at a time when the UK had no clear authority on the relevance of the principle of “fair basis” to the UK Patents Act 1977 which no longer referred to that concept, instead following Europe’s requirement for “support”.

This raises a further point – that there is nothing new, unusual, or necessarily improper in amending a claim to create an “intermediate generalization” per se. The European authorities have grappled with intermediate generalizations in numerous decisions and they have repeatedly determined that some intermediate generalizations are allowable and some are unallowable – and that the determinant will be whether the specification discloses additional technically relevant information as a result of the amendment.⁹

Les Laboratoires Servier v Apotex

First, the sole authority relied on by Emmett J in Les Laboratoires Servier v Apotex was Lockwood v Doric and neither the High Court in that proceeding nor Emmett J in his decision used the expression “intermediate generalization”. Moreover, Lockwood v Doric is widely regarded as setting a low bar to considering what is fairly based.

Second, Emmett J’s discussion did not form part of the ratio of the Full Court in Les Laboratoires Servier v Apotex, instead the majority (Kenny and Stone JJ) stated that the primary judge was correct to not exercise her discretion to allow any amendment, and they hence did not consider the criteria for allowability of the particular amendment that was proposed. Emmett J’s separate judgment did consider the allowability of the amendment and took a narrow view that the new feature was inextricably linked back to other features of the invention and that seeking to isolate it should not be allowed.¹⁰ In line with the majority, Emmett J additionally decided that the primary judge was correct to not award the discretion to allow the amendment.

Third, the primary judge in Apotex v Les Laboratoires Servier¹¹ followed the well-established principles of the Full Federal Court (in particular) that “in substance disclosed” is similar to “fair basis” or possibly even more liberal. That extensive higher court authority for the law on fair basis, in substance disclosure and amendment allowability has been sufficiently well developed under Australian law that reference to UK and EP jurisprudence by Beach J in obiter should, respectfully, carry no weight.

Possible consequences

Despite our analysis, Beach J’s speculation that the “new” law on amendment allowability may apply as strictly to the “old” law is troublesome since many of the granted patents currently in force in Australia were subject to the pre-Raising the Bar version of the Patents Act 1990, and those patents may stay on the register for another decade or more.

If amendment of any one of the originally filed claims was previously allowed by the Patent Office, but is later deemed unallowable by a Court, the priority date of that claim will be deferred¹² which may lead to invalidation of the patent for lack of inventive step or (curiously) even lack of novelty.

That was precisely the outcome in AstraZeneca v Apotex¹³ where the majority held that: an amended claim was “fundamentally inconsistent”¹⁴ with the original disclosure; that the amendment should therefore not have been allowed; that, as a result, the priority date of the amended claim should be deferred; and that the claim lacked novelty. Relevantly, the parties to the decision of the Full Court in AstraZeneca both directed¹⁵ the Court to the liberal test for “fair basis” as determined in Lockwood v Doric.

It is not difficult to imagine a similarly fatal outcome for a large number of patents in which the Patent Office has allowed an “intermediate generalization” amendment which could be tested in invalidity proceedings under Beach J’s suggested criteria for amendment allowability.

In our view, Beach J’s speculation in CSIRO v BASF Plant Science GmbH should not (but may inevitably) provide fertile ground for any defendant to revisit the amendment history of a granted patent with an eye to applying stricter requirements for amendment allowability, so as to invalidate that patent.

This is precisely the logic that was applied in a recent decision of the Australian Patent Office in Merck & Cie,¹⁶ where a very experienced Deputy Commissioner of Patents conducted a hearing following re-examination of an application which was subject to the pre-Raising the Bar law. The sole focus of the hearing was to determine whether certain amended claims defined a non-obvious invention. However, after determining that the amended claims did not lack an inventive step,¹⁷ the Deputy Commissioner (in obiter¹⁸) also cited the obiter reference in CSIRO v BASF Plant Science GmbH which was to the obiter reference in Les Laboratoires Servier v Apotex Pty Ltd as basis for applying the law on “intermediate generalizations” to the amended claims being re-examined. The Deputy Commissioner expressed “concerns” that the original specification did not provide a “specific” enough disclosure of the combination of features in the amended claims for them to be fairly based. This approach seems to favour Emmett J’s obiter¹⁹ over the full weight of the High Court in Lockwood v Doric²⁰, and may prove troublesome for the patentee.

In our view, the compounding of non-binding obiter is unhelpful to the development of coherent law, and this decision validates the authors’ concerns for future developments in this area.

The appropriate standard

We argue that Beach J’s suggestion that the standard for amendment allowability has always been strict is not appropriate.

Before the Federal Court of Australia, James & Wells successfully appealed against a decision of the Australian Patent Office to disallow a set of complex amendments in United States Gypsum Company v CSR Building Products Ltd²¹. On one interpretation the amended claims could be characterized as intermediate generalizations. Moshinsky J applied established authority in asking “whether there is “a real and reasonably clear disclosure” in the relevant specification of what is claimed in the amended claim, such that “the alleged invention as claimed is broadly, that is to say in a general sense, described in the body of the specification””.²² The amendments were allowed.

Very recently James & Wells appeared again in the Federal Court of Australia before Besanko J in Boehringer Ingelheim Animal Health USA Inc v Elanco New Zealand,²³ this time defending an appeal against a decision of the Australian Patent Office to allow amendment of a set of claims to introduce a disclaimer of matter not explicitly referred to. At its heart the proceeding was probing whether such a disclaimer “added matter” and whether the claimed invention was “fundamentally inconsistent” with the invention originally described. The decision in AstraZeneca v Apotex was relied on heavily by both parties – a decision in which Besanko J was a member of the Full Court’s bench. We look forward to Besanko J’s further clarification of the law of amendment allowability.

In the recent Full Federal Court decision in Meat and Livestock Australia Limited v Branhaven LLC,²⁴ the Full Court did not cite any of the three authorities discussed above that Beach J referred to in CSIRO v BASF and instead indicated that (emphasis added):

All other things being equal, a claim that defines an invention in terms that are narrower than a more general description in the body of the specification would support is not likely to travel beyond what is more generally described. But there may be some situations in which what is more specifically defined results in a claim that travels beyond what is described in the specification… In these situations a claim may be invalid if the invention more specifically defined is an invention that is different from the invention described in the specification as opposed to some narrower embodiment of the latter.

Conclusion

With the greatest of respect to an accomplished Intellectual Property Justice of the Federal Court of Australia, we submit that there is a difference in scope between the previous version of s 102(1) and its current form, and that the distinction is not subtle. Rather:

• the pre-Raising the Bar version of s 102(1) requires that there simply be a “real and reasonably clear disclosure” of the matter introduced as a result of the amendment and that it should not “travel beyond” the invention originally described. The provision “should generally be given a liberal construction”²⁵ and that the phrase “not in substance disclosed” should be interpreted in “a broad common sense way”.²⁶ There should be a focus on whether the amendments result in an invention that is “different from”,²⁷ or “fundamentally inconsistent”²⁸ with the originally described invention; as against
• the post-Raising the Bar version of s 102(1) which provides a “strict” requirement for a “clear and unambiguous” disclosure of the matter introduced as a result of the amendment unfettered by any “context” which would prohibit its isolation from that context.²⁹

As always, seeking advice from experienced patent attorneys who have substantial prosecution and litigation experience will help you to avoid the unintended consequences of what is otherwise a “simple” amendment. We look forward to talking with you.

In our next article we discuss how these developments might impact on New Zealand law.