Challenges Post­ Selex

Twenty­-five years ago Tuerk and Gold reported the Selex method. Being the first European patent attorney to prosecute patents relating to aptamer/SELEX technology, Richard Clegg discusses some of the challenges at the EPO today when attempting to obtain adequate class-­level protection for functionally defined molecules.

Occasionally, advances in biotechnology yield novel classes of molecule. Monoclonal antibodies represent an early example. Never patented at a class level, structurally defining monoclonal antibodies as a class would have been challenging against the backdrop of polyclonal antibody prior art available at the time—but at least antibodies have common structural characteristics that distinguish them from other substrate-­binding proteins.

The Selex method

Over the years several classes of nucleic acid molecule have been identified by developing screens for functional characteristics. For example, in recent years screening for transcriptomes has generated structurally diverse pools of nucleic acids having a common functional property. Other examples are synthetic RNA and ribozymes.

In 1990, Craig Tuerk and Larry Gold reported one such method, calling it Selex (Systematic Evolution of Ligands by EXponential enrichment). It identified a new class of nucleic acid molecule that exhibits high affinity binding to a selected target molecule. In doing so they opened up a new field of research, which has led to new diagnostic and therapeutic products.

The nucleic acids identified by their method were originally referred to as nucleic acid ligands but for many years now have been referred to as aptamers. Aptamers are short nucleic acids characterised by high affinity binding to the selected target. The nucleotide sequence of an aptamer for a target is not predictable and two aptamers to the same target often have unrelated sequences.

To obtain a product (composition of matter) claim at the European Patent Office (EPO), novelty must of course be established. During novelty examination the examiner will consider whether the claim is clear enough for it to be distinguished over the known prior art. When claiming products, a molecule of a known structural type that is characterised by function alone is unlikely to be considered clear enough, particularly where the prior art contains structurally similar molecules not described, or tested, in respect of the function being used to define the claim.

The EPO’s approach

Today, antibodies can be an exception but it has taken considerable time for the EPO to reach this position and an antibody analogy can be difficult to deploy if the molecular class at hand does not have a similar distinct structural characteristic.

For example, a claim to a short nucleic acid characterised by binding to target Y, or having gene-repressing ability, will be assessed against the backdrop of a very large number of short nucleic acids in the prior art for which such functions have not been tested.

The EPO can be expected to consider that at least one such molecule in the prior art may inherently have the property at hand and, where that probability is not considered to be de minimis, can be expected to take the view that the claims are not clear enough to allow novelty to be established and/or that an undue burden is placed on a third party to determine whether a nucleic acid in their possession in fact has the property and therefore falls within the claim, ie, insufficient certainty to determine infringement.

Introducing gross structural claim limitations, eg, non-­natural chemical modification, can reduce the extent of prior art to some degree but it can still be a considerable body of art and so the problem may remain.

So while pursuing class level product claims before the EPO is likely to be part of the strategy, it may be very difficult, even impossible.

In light of the difficulties in pursuing class level product claims before the EPO, what else can be done?

Product-by-process claims

Product-by-process claims are allowable before the EPO. Although not preferred they may be accepted where there is an inability to define the product satisfactorily by reference to composition, structure or other testable parameter, and to deny patent protection would be unfair in view of the contribution made to the art. The product itself must still be new and, therefore, must also be clearly defined under article 84 of the European Patent Convention (EPC).

The boards of appeal have acknowledged that the process part of defining the product may be used to provide clarity of the claim but the balance of any burden placed on the EPO or third parties to determine whether a product (in the prior art or after the filing date) falls within the claim will be made against the desirability to provide fair protection to the patentee to recognise the breadth of contribution made to the art. The patentee can come off second best.

Functionally defined products are often identifiable by a form of identification/screening method, probably performed in vitro. Patenting that method before the EPO will normally be reasonably straightforward. Where product claims are difficult to obtain, method claims can be valuable. This is because the EPC provides protection for products that are “directly obtained” by a patented process. This protection is, however, of reasonably narrow scope.

The ‘product’ obtained by the method has been construed quite narrowly and a distinction may be made between the crude product of an in vitro screening method and the refined commercial product, the latter not necessarily being “directly obtained” by the patented process.

A related problem is that in the field of nucleic acids, one ‘product’ of the screening method is sequence information, which can be rapidly disseminated across the globe and a new ‘product’ synthesised according to that sequence. Not only does this make reliance on the “directly obtained” provision more complex, but it also raises the issue of infringement of method claims between jurisdictions, eg, method performed in country X and new product synthesised according to sequence information obtained from that method in country Y, possibly some years later.

Refining the method claim can be considered, eg, to characterise the method as one of producing a commercial grade product. The EPO might consider this to drift into reach¬-through territory, but for a unified class of molecule this problem can sometimes be avoided by noting that the products obtained from the screening steps for formulation in the production steps are not completely open--ended but must be of the class.


Framing your claims

Another approach used successfully is to frame the claim in terms of production of the commercial grade product starting from sequence information, ie, a claim to synthesis of the product according to sequence information obtained from the screening method.

Such claims can attract rejection for lack of clarity. Over the years the EPO has moved towards citing the Enlarged Board of Appeal’s decision in G2/88 to formulate an examination guideline providing the basis to reject such claims as combining irreconcilable types of process. This approach seems open to question, given that the comments in G2/88 would, by common law standards, be considered obiter and not addressing this issue and also clearly state that “there are no rigid lines of demarcation between the various possible forms of claim”.

“If the novel molecular class at hand can be defined as having a structure that corresponds with the primary functional property, eg, target binding, the antibody analogy may be deployed."

At this point it is worth noting the particular position taken by the EPO towards product claims for antibodies. For this class of molecule the EPO has become comfortable with the possibility of defining the antibody by providing a functional characterisation of the antibody and this is sufficient where the antigen is new. Where the antigen is known, a further functional characterisation of an antibody sub¬genus can be accepted by the EPO, eg, a functional effect achieved by an antibody of the sub¬genus when bound to the antigen.

The EPO appears to have reached this position because it has become comfortable with the idea that an antibody per se can be structurally distinguished from other proteins, such that the requirement to afford third parties reasonable certainty regarding their ability to test for antigen binding and any additional functional property recited in the claim may be considered appropriately balanced against the need to afford the patentee fair protection for the invention. That said, this is not a given and even in the antibody field the outcome can be expected to depend on the facts.

Therefore, if the novel molecular class at hand can be defined as having a structure that corresponds with the primary functional property, eg, target binding, the antibody analogy may be deployed, and certainly should not be overlooked.

A special type of product claim

I have formulated and deployed strategies of the kind described above over the last ten years. With sufficient effort useful outcomes are possible. Obtaining composition of matter claims will remain an important goal and it is unnatural for many applicants to contemplate that it may not be possible. But there is another way.

This of course is the EPO medical use claim for first or later medical uses. Since the revision of the EPC, both claim types can be thought of as a special type of product claim, limited by its purpose of treatment (or diagnosis) when performed on the human or animal body. Applying this claim format can make a considerable difference to the problems associated with the ‘pure’ product claims discussed above. The reason is that the intention of the user becomes a relevant factor. That is, when considering the prior art the mere existence of a product of the same structural class, eg, short nucleic acid, should not be sufficient to raise a novelty concern unless it is indicated for the same medical use.

The functional characteristic, eg, target binding, may also be recited in the claim to add clarity over any prior art that does recite a medical use, and finally a product-by-process definition may be added to the claim to tip the balance of risk to third parties of uncertainty from a potentially unclear claim against the desire to afford the patentee reasonable protection for the invention.

The result can be a granted claim of broad therapeutic scope and where the products are intended for therapeutic use this may afford meaningful and valid protection in an acceptable prosecution timeframe. Of course, the product claims can still be pursued (as can in vitro method claims, eg, for diagnostic uses) but it may be sensible to do this via a divisional application strategy so as not to delay the issue of enforceable claims.

Later applications may initially identify sub¬genera of the functionally defined class and it may be possible to take a similar approach in such cases (and indeed it may become easier to deploy the antibody analogy and get product claims allowed), but as the technology progresses it is reasonable to expect that specific products for specific targets and/or individualised disease treatments will be identified, for which the normal approach to protection of the products and uses should apply.

All of the above is worth considering, if possible. No doubt, many readers will be familiar with the EPO’s strict and user-unfriendly approach to added matter. Unless language for the claim options discussed above appears in the European patent application as filed (for many readers, this will be the Patent Cooperation Treaty application) simply writing admissible claims of this kind may be impossible, leading to major frustration and expense.

Applicants who consult their European patent attorney at the drafting stage to seek incorporation of suitable language for the appropriate patent style claims should be rewarded when European prosecution starts.

Richard Clegg started working on aptamer/SELEX technology in 2003 and became the first European patent attorney to prosecute patents relating to this technology. He has since prosecuted in excess of 150 further patent applications for aptamer-related inventions.