You are here: Home > News > When is a Selection not a Selection?

When is a Selection not a Selection?

14 February 2021

One of the requirements for an invention to be patentable before the EPO is that the invention is novel. Novelty over a prior art disclosure is commonly conferred by the presence of at least one feature which is not disclosed in the relevant prior art document. However, this is not the only method. For example, it is sometimes discovered that particular members of a known, more general category possess surprising or remarkable properties. It may be that only limited testing of the members of the category was performed in the prior art, and this failed to reveal the unique behaviour of certain other members. This is not uncommon, especially where the prior art document is a patent or patent application.

A particular example of this situation relates to numerical ranges, e.g. where it has been found that operating within a particular portion of a prior art range conveys surprising advantages. This situation most commonly arises in the fields of life sciences and chemistry. However, the approach taken by the EPO with respect to this situation is the same, regardless of the field of invention.

The Two-List Selection Principle – Discrete Lists

If, in order to arrive at the claimed invention, one is required to choose a single element from a single list of discrete elements in the prior art, the EPO does not consider this to provide novelty. In essence, this selection is not deemed to be sufficiently ‘hidden’ within the prior art for it to be considered a new invention. However, if two selections are required, this increases the number of permutations from which the skilled person must ‘sift out’ the invention. As such, the EPO considers that two or more selections warrants novel status because the claimed combination of choices was essentially hidden within the numerous possibilities of the prior art disclosure.

There are, however, some caveats.

The EPO requires the lists in question to be “of a certain length“. A selection invention is essentially an uncovering of a remarkable species (or group of species) hidden within a large array of permutations. Therefore, if the number of permutations is not sufficiently large, it is hard to argue that a particular member was genuinely ‘hidden’.

The two (or more) selections must also be “arbitrary“. If the prior art contains pointers that guide the reader towards making particular selections, then clearly the embodiments encompassed by the selections are not truly ‘hidden’. For instance, if all the examples described in the prior art document require the same selection (or if a specific member of a list is described as “preferred”), such a selection is likely to be disqualified.

Ranges – Continuous Lists

Numerical ranges are treated in a similar fashion. However, rather than a list of discrete integers, numerical ranges represent a continuous spectrum of numerical values between two end points. In order for a sub-range (or a particular value within a range) to be novel, the EPO requires that: the selected sub-range is narrow compared to the known range; and the selected sub-range is sufficiently far removed from any specific examples disclosed in the prior art and from the end-points of the prior art range.

Furthermore, the EPO considers whether the operation of the invention within the narrow sub-range produces some technical effect not described in the prior art, and regards this as confirming the finding of novelty.

If such a technical effect is present in the claimed sub-range, this can of course also help to satisfy the requirement of inventive step.

Overlapping Ranges

There are also cases where a claimed range may overlap with a prior art range. This happens not just with numerical ranges but with generic chemical structures too. In order to establish novelty in such a case, it is essential to demonstrate that the skilled person would not, taking into account the whole disclosure of the prior art document, reasonably consider working in the area of overlap. To demonstrate this, one typically needs to find in the prior art document a teaching away from the area of overlap.

Many of the standards employed when considering selection inventions are very much decided on a case-by-case basis. However, the underlying principle is that, if a genuine and surprising improvement is hidden with the sweeping ranges and statements of an earlier disclosure, there should be no barrier to novelty.


Nicholas Jones
Life Sciences & Chemistry

This publication is a general summary of the law. It should not replace legal advice tailored to your specific circumstances.

© Withers & Rogers LLP February 2021