Patent Law Update - ToolGen Incorporated v Fisher (No 2) [2023] FCA 794

ToolGen is a Korean company involved in the development of genome-editing products and services. This case is centred on ToolGen’s patent application for CRISPR gene editing technology. ToolGen’s patent was successfully opposed at the Patent Office by a “straw man”, with the Delegate rejecting the patent on various grounds. This case concerns ToolGen’s appeal of that decision (and the respondents’ cross-appeal in relation to grounds rejected by the Delegate).

After a comprehensive consideration of the relevant technical background and expert evidence, in this decision Justice Nicholas ultimately found:

• That the asserted priority document (P1) did not disclose and enable the claims of the Patent Application in suit and so none of the claims were entitled to the earliest priority date (being 23 October 2013) (s 43(2A)).
• The claims of the Patent Application would be invalid (if granted) because:

1. the specification of the Patent Application did not provide an enabling disclosure of the invention defined by the claims (s 40(2)(a));
2. claim 19 lacked clarity (s 40(3)).
3. Given the shift in the priority date, all of the claims lacked novelty or were obvious (s 18(1)(b)).

ToolGen has until 11 August 2023 to put on an application for amendment to the Patent Application (which the respondents have already indicated they will oppose). Notably, ToolGen had proposed that the matter be remitted to the Patent Office for consideration of any amendment application, but his Honour’s view was that, given the complexity of the matter, that would not be “desirable”.

The technology

CRISPR technology is revolutionary. Without delving too deeply into the molecular biology, this technology enables DNA to be cut and edited with unprecedented precision. The potential therapeutic implications of such technology has created great excitement throughout the scientific and wider community.

The Patent Application concerns a particular type of CRISPR technology termed the “Type II Clustered Regularly Interspaced Short Palindromic Repeats/Cas system” or “CRISPR/Cas system”. This system is best understood through the following image which is extracted in the judgment (and appears in P1 and the Patent Application itself):

The yellow circle represents the protein “Cas9”, which is an endonuclease that, in the context of the specification, was derived from the bacterial species S. pyogenes. When Cas9 forms a complex with a guide RNA (made up of crRNA (red box) fused to tracrRNA (blue box)) Cas9 cleaves (cuts) the target DNA (green box) at a specific site. In this image, the specific site is indicated by the two black arrows within the green box. Cas9 is “told” where to make this cut by recognising the sequence of nucleotides shown in the orange box (known as the “protospacer adjacent motif” (PAM)).

The specification of the Patent Application explains that the system can utilise the Cas9 protein itself or a nucleic acid encoding the Cas9 protein (e.g., DNA which would later be transcribed in the cell using the cell’s internal machinery to create the Cas9 protein). Likewise, and relevant to a key issue in dispute, the specification explains that the invention can utilise “naked” guide RNA (which had been transcribed outside the cell (in vitro) or a nucleic acid sequence which would be transcribed inside the cell (in vivo) into guide RNA. The specification explains that the use of naked guide RNA is preferable.

The person skilled in the art (PSA)

There was a dispute between the parties as to whether the PSA included (as part of the team) a microbiologist.

ToolGen put on evidence from a molecular biologist (Associate Professor Firestein) and a microbiologist (Professor Giffard) and asserted that the PSA was a team including both a molecular biologist and a microbiologist. The respondents put on evidence from two molecular biologists (Professor Thomas and Associate Professor Herold) and argued that the PSA did not include a microbiologist. Justice Nicholas looked at the disclosures in P1 and the Patent Application and found that, because P1 did not refer to using Cas9 from different bacterial species (and the Patent Application did) the PSA for P1 excluded a microbiologist whereas for the Patent Application, the skilled team included a microbiologist.

In reaching this view, Nicholas J considered a question which often arises in patent cases – whether a reference to another document constitutes incorporation by reference. ToolGen argued that the article by Jinek and others (Jinek), referred to in P1, was relevant to the question of identifying Cas9 from other bacterial species and this was one of the reasons why the PSA team included a microbiologist.

In finding that Jinek was not incorporated by reference into P1 or the Patent Application, his Honour referred to Lockhart J in Nicaro Holdings Pty Ltd v Martin Engineering Co [1990] FCA 40 in which his Honour said (at 549) that it is permissible to rely on a document incorporated by reference “provided that it is plain that the incorporation by reference unequivocally and plainly demonstrates that the draftsman has adopted the cross-referencing system solely as a shorthand means of incorporating a writing disclosing the invention”. Justice Nicholas considered that the reference to Jinek in P1 fell well short of meeting that test.

Construction

Two key issues of claim construction drove Justice Nicholas’ subsequent findings.

First, there was a dispute as to whether the phrase “a nucleic acid encoding a guide RNA”:

• was limited to DNA (or viral RNA) which was transcribed into guide RNA in vivo (the respondents’ argument); or
• referred to a nucleic acid sequence which enabled the guide RNA to perform its function (i.e., the guide RNA itself) (ToolGen’s argument).

This distinction was key to determining whether claim 19 lacked clarity. Independent claim 10 claimed “a nucleic acid encoding a guide RNA” and claim 19, which was dependent on claim 10, added the requirement “wherein the nucleic acid encoding the guide RNA is in vitro transcribed RNA”. If claim 10 was limited to circumstances where the guide RNA was transcribed in vivo then claim 19 lacked clarity because it was known that in vitro transcribed RNA (i.e., naked RNA) introduced into the cell could not be transcribed to guide RNA in vivo. ToolGen said that the respondents’ construction would result in: (1) excluding a construction of claim 10 which would enable claim 19 and claim 10 to coexist; and (2) a surprising scenario where the preferred form of the invention (i.e., the use of naked RNA) was not encompassed within the claims.

Justice Nicholas rejected ToolGen’s arguments and found that the respondents’ construction was consistent with how molecular biologists understood the word “encoding” (i.e., a code for transcription). His Honour also noted at [130] that “a patentee may have a good reason for introducing a seemingly inexplicable limitation”. Thus, his Honour concluded that claim 19 was unclear.

Secondly, the parties disagreed on whether “paired Cas9 nickases” fell within the scope of the claims. The respondents sought to argue that the claims included nickases as another basis for arguing that the claims were not entitled to their priority date (since P1 did not disclose nickases). Nickases are enzymes that operate in pairs such that there are two separate Cas9 proteins, each which its own guide RNA. Each Cas9 complex cuts a single strand of DNA in a different location remote from one another (resulting in which is called “sticky ends” of DNA).

ToolGen argued that the claims did not include nickases because all claims required a “double stranded break at a target nucleic acid sequence” and two breaks at remote locations resulting in sticky ends did not meet that definition. Justice Nicholas agreed with ToolGen that the claims did not include nickases but not for the all the reasons argued by ToolGen. His Honour considered that the use of singular terms in the claims excluded a Cas9 pair. His Honour was not persuaded by the respondents’ reference to the inclusive definition of “comprising” and “comprises” in the specification, because his Honour considered that this was a circumstance where “the context requires otherwise”.

The disclosure requirement

The law

This decision adds to the growing body of Australian case law aligning Australia’s section 40 requirements to the position in the UK and Europe. Relevantly, Justice Nicholas stated:

• At [166] and [175]: The test under s 40(2)(a) and section 43(2A)(b) is essentially the same. Both require the invention to be disclosed in a manner that is clear enough and complete enough for the invention to be performed by a person skilled in the relevant art. Nothing turns on the fact that s 40(2)(a) does not express the obligation by reference to the invention “in the claims”, the obligation imposed by this section relates to the invention as claimed.
• At [168]: “It is apparent that the amendments to s 40(2)(a) of the Act were made with the intention of aligning the Australian law of sufficiency with UK and European law. In particular, the old law, which had generally been held to require the enablement of only a single embodiment of the invention within each claim, was done away with.”
• At [193]: Notwithstanding the differences between statutory language in the UK and EPC provisions and section 40(2)(a), the position under Australia law “is not materially different from UK law as explained in the English authorities to which I have referred”.
• At [183]: Contrary to ToolGen’s submission, the absence of the words “in respect of the same invention” (which appear in Art 87(1) of the EPC) in s 43(2A)(b) does not mean that s 43(2A)(b) imposes no requirement that the priority document and the patent application be in respect of the same invention. “[T]he invention of the claim must be disclosed in the prior application and must be also disclosed in a manner that is clear enough and complete enough for the invention in the claim to be performed by the person skilled in the art.” ToolGen had argued that an important distinction between Australian and UK/EPO law is that Australian law does not require strict disclosure of the “same invention” in the priority document, it only requires enablement. His Honour’s rejection of this submission might come as a surprise (and concern) to Australian patent attorneys as it appears to significantly elevate the requirements for priority documents.
• At [192]: Justice Perram’s statements in Encompass Corporation Pty Ltd v InfoTrack Pty Ltd [2018] FCA 421 regarding section 40(2)(a) are authority for the proposition that section 40(2)(a) does not require that the complete specification to make the nature of the invention plain, but that finding should not be taken as suggesting that there is no requirement under section 40(2)(a) that the invention be disclosed.

Priority date – does P1 provide an enabling disclosure?

The respondents identified five areas in which they argued that the claims of the Patent Application were not enabled across their breadth by the disclosure in P1. In this portion of the judgment his Honour went into considerable detail setting out the expert evidence relevant to each area, analysing whether the matter identified was disclosed in P1 and if not, considering whether the PSA could perform the invention without undue burden. The extent of the expert evidence and analysis required in this portion of the reasons for judgment signals that, in addition to inventive step, the disclosure requirement under Australian law is fast becoming another area which will generate significant legal expense. To illustrate his Honour’s approach to this aspect of the case we have outlined his Honour’s findings on two of the five issues.

A system from a bacterial species other than S. pyogenes

The independent claims of the Patent Application are not limited by bacterial species from which the system was derived. Therefore, the Court had to determine whether P1 disclosed a system for cleaving DNA using Cas9 derived from a bacterial species other than S. pyogenes (which was plainly disclosed in P1). ToolGen argued that P1 disclosed any Cas9 from a Type II CRISPR system capable of forming an active endonuclease when complexed with guide RNA. Justice Nicholas rejected that argument because he did not consider P1 to identify any principle of general application which would permit the skilled addressee to determine whether any particular Type II Cas9 might work.

Despite concluding that P1 did not make the relevant disclosure (the relevant disclosure being, on his Honour’s view of the law as set out at [183] of his reasons, the “same invention” as claimed in the Patent Application), his Honour went on to consider, in the event he was wrong on that point, whether the disclosure in P1 was clear enough and complete enough for the invention to be performed by the PSA. After setting out the steps each expert said would be required in order to identify an alternative bacterial species, including in the event that he was wrong about a microbiologist not being part of the skilled team, his Honour ultimately concluded that it would have been apparent to the skilled team that there was considerable uncertainty as to whether or not a system derived from any other bacterial species would work in eukaryotic cells. Further, his Honour found that this would involve a significant, multi-step research project that would not be straightforward or routine. Accordingly, his Honour found that there was no enabling disclosure by P1 in this respect.

Nuclear localisation sequences (NLS) and their location

A NLS is a “tag” that when added to a protein enables it to be imported into the cell nucleus. ToolGen argued (in a similar vein to its argument on an alternative bacterial system) that P1 implicitly discloses the use of any NLS capable of mediating the entry of the Cas9 protein into the nucleus and is not limited to the tag expressly disclosed. In contrast to his Honour’s finding on bacterial species, Nicholas J found that the work associated with the use of a different NLS would be routine and straightforward and would not create undue burden. In this regard his Honour considered the disclosure in P1 analogous to a description of an embodiment of a mechanical device in which a screw is used to fasten two components. His Honour remarked that, to the PSA, this could be understood as a disclosure of not only a screw but other forms of suitable fasters such as a bolt or rivet.

Does the Patent Application provide an enabling disclosure?

In contrast to his analysis of P1, his Honour’s analysis of this ground was very brief. Ultimately, his Honour found that the Patent Application did not provide an enabling disclosure for effectively the same reasons as articulated with respect to P1 (save for in relation to arguments not pressed by the respondents in relation to the Patent Application).

The support requirement

The law

Consistent with his approach on the disclosure requirement, Nicholas J adopted Justice Burley’s view in Merck Sharp & Dohme Corporation v Wyeth LLC (No 3) [2020] FCA 1477 that the approach by Aldous J in Schering Biotech Corp’s Application [1993] RPC 249 at 252-3 broadly encapsulates the support obligation under section 40(3).

His Honour made the following observations about the operation of the support requirement (emphasis added):

• At [396] “The monopoly, according to UK and European authorities, must be justified by the technical contribution to the art that arises from the disclosure of the specification.”
• At [397]: “The practice of claiming inventions that are not shown to have a sufficiently plausible or credible justification or support is sometimes referred to as speculative claiming.”
• At [400]: “It can be seen that the concept of plausibility has been developed in the UK authorities as a check on speculative claiming and to ensure that the patentee’s monopoly is no more extensive than the contribution to the art made by the relevant disclosure.”

These statements implicitly suggest (as we proposed in our recent article here) that the UK concept of “plausibility” might soon find itself part of Australian law by way of section 40(3). While interesting, unfortunately the issue of plausibility did not arise in this case because his Honour found that the respondents did not assert lack of plausibility on the face of the Patent Application. Instead, the respondents argued that the Patent Application did not disclose any principle of general application.

Justice Nicholas’ finding that the Patent Application lacked support followed his finding on lack of enabling disclosure. His Honour concluded that while a claim can meet the requirements of section 40(2)(a) by providing an enabling disclosure but not meet the support requirement in section 40(3), it is difficult to see how a claim to an invention for which there is no enabling disclosure could meet the support requirements. His Honour concluded at [410] that “[i]n such circumstances, the scope of the monopoly defined by the claim could not be justified by the technical contribution to the art. The two requirements are closely interrelated and not wholly distinct in their fields of operation.”

Novelty

ToolGen accepted that if claims 1-8 and 10-18 were not entitled to priority from P1, then those claims lacked novelty. With regard to claims 9, 19 and 20, his Honour found that if, contrary to his findings, those claims included the use of in vitro transcribed guide RNA (and claim 19 did not lack clarity), those claims would lack novelty in light of the publication “Wang 2013”.

Inventive step

As with novelty, ToolGen accepted that if claims 1-8 and 10-18 were not entitled to priority from P1, then those claims were obvious.

As to claims 9, 19 and 20, if, contrary to his findings, those claims included the use of in vitro transcribed guide RNA (and claim 19 was clear), his Honour found that the evidence showed that those claims would have been obvious in light of the common general knowledge and three separate prior art documents (considered separately). Further, on his Honour’s construction of claims 9 and 20 (i.e., that the claims are limited to the use of a nucleic acid such as plasmid DNA encoding guide RNA) he found that those claims lacked inventive step in light of the common general knowledge and two prior art documents (considered separately), each of which described experiments where the guide RNA is encoded by plasmid DNA.

Finally, Nicholas J also found the last remaining claim, claim 21, obvious. Claim 21 introduces the additional step where the nucleic acid encoding the Cas 9 protein is introduced to the cell before introducing the nucleic acid encoding the guide RNA. His Honour found that the Patent Application did not provide any indication that claim 21 added anything that would render claim 21 inventive, and that it would have been obvious to the PSA that the method in claim 10 could be performed in a stepwise fashion, as described in claim 21.