The use of product or sponsor codes in clinical trial related documents is common practice in the pharmaceutical space. However, such codes may not always be sufficient to disregard a publication as relevant prior art against patent claims.
In recent decision T 1255/21, the EPO Board of Appeal finds that the use of a product code or sponsor code for a compound or composition in a clinical trial protocol does not affect the status of the protocol as prior art if the skilled person would have been able to identify the compound or composition based on the description of the components of the product code combined with their common general knowledge. This was the case despite the product code not being directly equated to a particular compound or composition in the art.
Background
The decision in suit relates to European patent 3140320 in the name of Targovax Solutions AS. EP3140320 claimed a co-therapy for the treatment of cancer, comprising a pyrimidine analogue administered alongside at least one peptide selected from a list of specific RAS mutant peptides.
A clinical trial protocol entitled "A Phase I/II Trial of TG01 and Gemcitabine as adjuvant therapy for treating patients with pancreatic cancer" registered with EUCTR (the EU Clinical Trials Register) and published by WHO was full prior art against the claims of the Patent. Gemcitabine is a known pyrimidine analogue. However, the product code "TG01" is not commonly known, and it was disputed between the parties whether the skilled person would have been able to determine what the composition identified as "TG01" actually consisted of.
At first instance the opposition division (OD) considered that the clinical trial protocol was not a suitable starting point for the assessment of inventive step because the protocol “does not disclose the sequences of the peptides and no results for the trial are shown”. The OD considered the granted claims to be both novel and inventive and rejected the opposition filed against the Patent.
The Appeal Board Decision
In contrast to the decision of the OD, the Appeal Board found the granted claims to lack an inventive step in view of the clinical trial protocol publication (hereinafter referred to as “the protocol”).
The Board considered that the composition disclosed in the protocol “even at first glance shares at least several features with the composition of the claim and is intended for the same purpose, namely ‘treating patients with pancreatic cancer’” and was thus a suitable starting point for the assessment of inventive step (section 3 of the Reasons).
Product code “TG01”
As a first point, the Appeal Board considered whether TG01 would be understood by the skilled person to encompass peptides of granted claim 1. Whilst TG01 was not “commonly known” and could not be directly equated to a particular compound or composition in the art, the Board found that its description in the protocol could be identified in disclosures forming part of the skilled person’s common general knowledge.
The protocol described TG01 as a list of seven components, each of which having a “sponsor code” and an “other descriptive name”. The Board found that “[f]rom this list the skilled person would understand that the product ‘TG01’ consists of seven components [relating] to the oncoprotein "p21 RAS" because this is part of each of the seven other descriptive names”. The Board also noted that “KRAS status and recurrence” is also mentioned in the context of the “exploratory objective” of the protocol (sections 6 and 7 of Reasons). The Board also noted that the skilled person would conclude from the protocol that TG01 consisted of peptides or proteins due to the requirement for a specific DTH skin test reaction which cannot be achieved with RNA, DNA or material in a cell (section 10 of the Reasons).
Noting that “[i]n a fast-moving field such as medical research, and in particular the field of oncology, review articles can be considered to reflect the skilled person's common general knowledge” (section 12 of Reasons, citing T 1110/03), the Board considered the disclosure of review articles in the field of cancer-related KRAS mutations as common general knowledge (CGK). Considering the CGK, the Board found that the descriptive names of the seven components of TG01 which include letter codes all corresponding to well-known one-letter amino acid codes (12-A, 12-C, 12-D, 12-R, 12-S, 12-V, and 13-D), correspond to the “seven most common KRAS mutations in positions 12 and 13” (section 12 of Reasons).
The Appeal Board concluded that “the use of the code-name ‘TG01’ in [the protocol] does not affect the status of this document as a realistic starting point for assessing inventive step because the skilled person would have been able to determine what the composition identified as ‘TG01’ consisted of.” (section 16 of Reasons).
Reasonable expectation of success
Consequently, the protocol was considered as a suitable starting point for the assessment of inventive step, with the Appeal Board concluding that putting the proposal of the protocol into practice offered the skilled person a reasonable expectation of success (sections 22-26 of Reasons).
The Board considered that the “only difference between the disclosure in the patent and the disclosure in [the protocol] is that the former discloses that the therapeutic effect which the trial is set up to test, is actually obtained”. The Patentee attempted to argue that “an improvement of the therapeutic effect could not be expected from the disclosure of [the protocol] thus rendering the combination treatment inventive”. However, the Board found that “[p]utting [the protocol’s] proposal into practice would inevitably have resulted in an improved therapeutic effect, if such effect was indeed achievable” and that “[s]uch improvement therefore represents a bonus effect, which according to established case law does not contribute to an inventive step” (sections 17 and 18 of the Reasons). Thus, the Board considered that “[t]he question to be answered in assessing the obviousness of the claimed subject matter is whether or not the skilled person starting from the disclosure of the clinical trial proposal in [the protocol] would have reasonably expected that putting it into practice would result in an effective treatment for pancreatic cancer patients” (section 22 of the Reasons).
The Patentee argued against a reasonable expectation of success. The Patentee pointed to a reference to the assessment of "the potential for interference of Gemcitabine on immune response to TG01" in the protocol. However, the Board considered that the reference to “interference” would have been understood by the skilled person to be “a standard indication for any combination therapy, and as such it would not impart to the skilled person any particular prejudice against the proposed clinical trial”. The Patentee further argued for a lack of reasonable expectation of success based on earlier clinical trials reporting that “a peptide vaccination combined with gemcitabine had been stopped because of a lack of improvement in patient survival”. However, the Board found that these earlier reports were linked to “a lack of effect of the particular vaccine”. In the present case, since both TG01 and gemcitabine were known to have a therapeutic effect in the treatment of pancreatic cancer patients at the relevant date of the patent, the Board found that there was no “prejudice or teaching away from combining peptide vaccines with gemcitabine in the art, which would have dissuaded the skilled person to put the clinical trial proposal of [the protocol] into practice” (section 26 of the Reasons). This is in line with previous decisions supporting the position that a clinical trial protocol provides an expectation of success that the therapy in the clinical trial would work, unless a person skilled in the art was dissuaded from this by the prior art (e.g. T 239/16, T 1123/16 and T 1806/18 (discussed in our article here)).
In view of the above, the Board found the claimed subject matter to lack an inventive step.
Concluding Remarks
This is not the first time that the Appeal Boards have considered whether a product code used in the prior art would prevent the skilled person from equating a claimed compound with one disclosed in the prior art. In decision T 1732/18, the prior art referred to the claimed compound only by an “internal project code name”. The Appeal Board found that “the person skilled in the art would have found no difficulty in looking up the chemical identity and preparation of [internal project code name] in the appellant’s further publications on this compound”. The Board noted that since “this was a recent development, only a limited number of publications would have had to be viewed” (section 9.4 of Reasons). A similar situation arose in decision T 2963/19 where the claimed compound was only referred to by a code-name in the prior art, but where “the skilled person was able to identify [the code-name] as a known and available product” (section 4.1.3 of Reasons).
Interestingly, in the present case, the specific product code “TG01” was not found to be defined verbatim in the prior art. As noted by the Board, TG01 was not “commonly known” and could not be directly equated to a particular compound or composition in the art. Instead, the description of each of the components of the product code in the clinical trial protocol together with the skilled person’s common general knowledge was used to identify the identity of the product code. Therefore, prudence should be used by patent applicants who seek to use a product code to sever the link between a claimed compound or composition and a compound or composition described in the prior art.
As with all decisions related to clinical trial prior art, each case must be assessed based on its own facts. However, these decisions indicate that the use of a product code for a compound or composition in a clinical trial protocol will not prevent its relevance against patent claims where the skilled person would have been able to determine the identity of the compound or composition based on the skilled person’s common general knowledge at the effective date of the patent.
Social Media cookies collect information about you sharing information from our website via social media tools, or analytics to understand your browsing between social media tools or our Social Media campaigns and our own websites. We do this to optimise the mix of channels to provide you with our content. Details concerning the tools in use are in our privacy policy.