Home Insights Virus filtration patent gets clean bill of health

Virus filtration patent gets clean bill of health

Read time
4  minute read
Date published
02 March 2022

In a recent decision of the Australian Patent Office,[1] a patent application directed to a membrane for filtering biological products was upheld following opposition.  Attacks made on various grounds failed, with the Delegate considering that the claims could be clearly construed.  The decision also provided useful guidance on ‘parameteritis’ and prior use.

Griffith Hack acted on behalf of the Applicant, Asahi Kasei Medical Co., Ltd. in the proceedings.

Background and Technology

The patent application (AU2015309939) concerns porous membranes which are used to filter and capture virus particles from biological products such as fractionated plasma products and biopharmaceuticals.  The claims of the application were directed to membranes having a particular distribution of pores within the membrane cross-section, which were associated with properties such as efficient filtration and removal of virus, together with reduced spoiling or clogging of the membrane over time.  Pore sizes and their distribution were determined by the use of scanning electron microscopy (SEM).

EMD Millipore Corporation opposed grant of AU2015309939 in 2018.  Following evidence rounds, and a procedural opposition concerning claim amendments which was decided in favour of the Applicant, a hearing was held in November 2021. Various grounds of opposition were pressed, with the main issues relating to claim construction and clarity, best method, and lack of novelty based on prior use of the Opponent’s Viresolve® Pro (VPro) membrane.

Claim construction and clarity

The claims of AU2015309939 were directed to a membrane having a pore structure defined by several parameters which required analysis by SEM, including characterising the membrane in terms of an asymmetric structure (where average pore size changes on moving through the interior of the membrane), coarse and dense layers, and a gradient index determined based on the average pore diameter in adjacent visual fields of the coarse and dense layers.

The Opponent argued that such terms were not clearly defined and that a person skilled in the art would not be able to identify membranes falling within the scope of the invention. However, the Delegate disagreed, understanding the membrane to have a dense layer defined by 1µm increments from the downstream side, the dense layer having a thickness equal to the number of 1µm visual fields that have an average pore diameter of 50 nm or less.  The coarse layer was the adjacent layer having 1µm visual fields with an average pore diameter of more than 50 nm.  Other features in the claims were then also resolved in favour of the Applicant.  In arriving at that conclusion, the Delegate referred to various passages in the description and examples, and was also assisted by expert evidence regarding the meaning of terms in the claims as supported by the specification.[2]

The Opponent also argued that the specification did not sufficiently describe how a sample should be prepared for SEM analysis, nor disclose the exact parameters to be set on the actual microscope. However, the Delegate found that a person skilled in the art would be “well equipped with objectively standard techniques”[3] and “have a common understanding of operating parameters… to create reproduceable and useful images of the scale of the present invention”.[4]

Prior use

The Opponent alleged lack of novelty based on prior use of their own commercially available VPro membrane.[5] However, the Delegate found that the evidence did not establish that the VPro membrane possessed all essential features of the claims. The Opponent used a different method of SEM analysis in their experiments, applying Euclidean distance mapping to determine pore size rather than the perfect circle method described in the specification. They also did not perform appropriate measurements to determine if the VPro membrane fulfilled other parameters in the claims, such as dense layer existence ratios and standard deviation of pore sizes. The Opponent argued that features of the claims were arbitrary parameters (discussed below), but the evidence was not sufficient to establish prior use.

The Applicant also asserted that the Opponent had not adequately established details of the alleged public disclosure, or that its membrane was reproducible by a skilled person.  Ultimately, the Delegate was able to settle the question of novelty without deciding on those points, but overall this decision serves as a reminder that the burden of establishing prior use is significant.  Care needs to be taken when preparing evidence of alleged prior use to address questions including: was there a public disclosure, if so when did it take place, and what exactly was disclosed to the public.


The Delegate also considered the issue of ‘parameteritis’ when assessing novelty. The term ‘parameteritis’ arose in the decision of Laddie J in Raychem Corp’s Patents [1998] RPC 31 and was described as a:

practice of seeking to repatent the prior art by limiting claims by reference to a series of parameters which were not mentioned in the prior art.[6]

The Opponent relied on this approach, arguing that certain parameters defined in the claims were “arbitrary”, did not achieve a technical effect and should be disregarded. However, the Delegate instead agreed with the Applicant’s position that the parameters defined in the claims combined to provide a technical advantage, could not be considered an “arbitrary convenience”, and were essential to the invention.[7] The Delegate also found that “parameteritis” is not a term found in the Patents Act, and does not represent a separate ground that a claim needs to meet.[8] Whilst at times a feature falling under the label of parameteritis may be inessential for the purposes of a novelty consideration (and only essential claim features need to be taken into account for assessing novelty under Australian practice[9]), in this case the specification made clear that the parameters were essential features.

Best method

A lack of best method attack was also raised, with the Opponent asserting that certain details were not provided in the specification relating to how SEM images of membranes were acquired.  However, the claims were product claims, and the Delegate held that the nature of the invention was not in an improvement in SEM imaging or image processing techniques, but rather was the creation of porous membrane filters with specific physical properties.  The best method enquiry was therefore directed to the creation of membranes, and the evidence did not establish that the applicant was aware of a better membrane product than those provided in the examples.[10]

How we can help you

Griffith Hack are pleased to support our clients on a range of matters from patent prosecution through to contentious matters such as opposition and litigation. We have an integrated team of patent attorneys and lawyers with strong technical and legal capabilities, and are committed to achieving the best possible outcomes for our clients.

If you would like to know more on how Griffith Hack can assist you, please get in touch.

[1] EMD Millipore Corporation v Asahi Kasei Medical Co., Ltd. [2022] APO 6

[2] Ibid, at [46].

[3] Ibid, at [54].

[4] Ibid, at [58].

[5] Ibid, at [117].

[6] Raychem Corp’s Patents [1998] RPC 31 at 37.

[7] Ibid.

[8] Ibid, at [116].

[9] Patent Manual of Practice and Procedure, 2.4.10.

[10] Ibid, at [83].