Illumina v MGI Part 1: Mr Justice Birss on sufficiency, DNA sequencing and chocolate teapots

The High Court decision in Illumina v MGI ([2021] EWHC 57 (Pat)) is the first to take an in depth look at the thorny issue of sufficiency since last year's Supreme Court decision in Regeneron v Kymab ([2020] UKSC 27) (IPKat). In the case in question, MGI's obviousness and insufficiency attacks on Illumina's DNA sequencing patents both failed. The obviousness case rested on the extent of the common general knowledge, and whether this would have motivated a skilled person to perform the invention. On sufficiency, Mr Justice Birss provided tests for determining whether an insufficiency issue amounts either to undue burden or lack of enablement across the whole scope of a claim (Regeneron sufficiency). Infringement will be covered in Illumina v MGI - Part 2. 

The Technology: Sanger sequencing

Sanger sequencing was a revolutionary DNA sequencing method invented in Cambridge in the 1970s (for an in-depth review of the history of DNA sequencing history, see Heather et al.). DNA is made up of chains of 4 nucleotides (A, C, G, T). Sanger sequencing uses modified versions of these nucleotides mixed in with normal nucleotides to synthesize copies of the DNA molecule to be sequenced. The modified nucleotides prematurely terminate DNA synthesis, producing DNA copy strands of different length. The modified nucleotides are also radiolabelled such that each terminated DNA strand will fluoresce according to whether the terminal modified nucleotide in the copy was a A, C, T or G. Measuring the nucleotide count and fluorescence of each DNA copy strand thereby enables the sequence of the DNA to be determined. For example, if A fluoresces "red" and C fluoresces "green" , and a "red" copy is one nucleotide larger than a "green", and sequence is AC. 

The patents at issue (EP 1530578 and its divisionals EP 3002289 and EP 3587433) related to improved modified nucleotides for use in Sanger sequencing. The claimed modified nucleotides could act as reversible chain terminators (RCTs). RCTs enable the blocking effect of the modified nucleotide to be removed once the identity and position of the modified nucleotide has been confirmed. The next nucleotide in the chain can then be added and the sequencing reaction continued. The RCTs claimed in the patents comprised an azidomethyl group. 

According to local legend, Sanger sequencing
was invented over a pint (or few)
in local pub, the Panton Arms

Obviousness

MGI argued that Illumina's patents were obvious in view of a journal article, Zavgorodny et al.. Zavgorodny was a paper from the field of organic synthetic chemistry which disclosed azidomethyl nucleosides (nucleotide derivatives). Importantly,  Zavgorodny did not mention anything to do with DNA sequencing or the use of RCTs. A key point of contention in the case was therefore whether the common general knowledge (CGK) of the skilled person would have included RCTs. According to MGI, a skilled person would know about RCTs, and it would thus have been obvious for a skilled person to use the RCTs disclosed in Zavgorodny to improve Sanger sequencing. 

Who was the skilled person and what did they know?

Crucial to the case was whether RCTs formed part of the CGK of the skilled person. To identify the skilled person, Mr Justice Birss summarised, it was necessary to ask what problem the invention aimed to solve and in which field of research this problem was located. It would be in this field that the skilled person or team should also be located (para. 58). 

There was little dispute that the problem to be solved was the provision of improved RCTs for use in DNA sequencing. The identity of the skilled team therefore included a molecular biologist/geneticist focused on DNA sequencing and an organic chemist (para. 99). In view of the expert evidence Mr Justice Birss was convinced that the CGK included scientific papers describing attempts at generating RCTs. However, at best, the CGK would have only amounted to a general understanding that these attempts had failed (para. 126). Importantly, the skilled person would not appreciate that the problem might be solved using alternative RCTs (para. 198). As such, the CGK as summarised by Mr Justice Birss effectively taught away from the claimed invention. 

Was it obvious?

Having set the scene, Mr Justice Birss then asked whether it would have been obvious for a skilled person to use azidomethyl nucleotides as RCTs. In some respects, Mr Justice Birss' characterisation of the CGK pre-empted this question. If the CGK did not include any motivation to try alternative RCTs to those already used in the failed attempts formed the CGK, it followed that it would not have been obvious for a skilled person to try using the molecules disclosed in Zavgorodny as RCTs (para. 195). Mr Justice Birss concluded that a skilled person would read Zavgorodny "and having done so, put it down and move on" (para. 199). The claims were thus found non-obvious. 

Sufficiency: Regeneron and chocolate teapots

In Regeneron, the Supreme Court had to decide whether a patent claiming a transgenic mouse product was enabled across the entire scope of the claimed invention. Crucially, a very commercially valuable product fell under the scope of the claim. However, by the time the case reached the Supreme Court it was accepted that a skilled person could not have made this commercial embodiment at the priority date. The Supreme Court decided that such a claim must fail for insufficiency.  

Chocolate teapot

In Illumina, MGI argued that Illumina's patent claims were insufficient, applying the principles of Regeneron. For Mr Justice Birss, the key question to answer was whether the disputed features of the claims were in fact relevant to Regeneron sufficiency. Mr Justice Birss highlighted that the only features requiring enablement over the entire scope of the claim under Regeneron are those relating to the core inventive concept of the claim.

By way of example, Mr Justice Birss considered a hypothetical patent claiming a new type of tea pot that does not drip (para. 277). The non-drip feature may be determined, for example, by the spout length and shape. The material from which the teapot was made would thus not matter for achievement of the non-drip feature. The teapot could be made from any conventional teapot materials. The patent claim for the teapot might therefore cover non-drip teapots made from any material, including future types of materials as of yet unknown. The claim would also cover teapots made from materials entirely unsuitable for teapots, such as chocolate. The material of the teapot is thus clearly important for allowing the teapot to function as a teapot. However, the material of the teapot also does not relate to the core inventive concept underlying the technical contribution of the non-dripping feature (which is determined instead by the spout length). As such, applying Regeneron, a non-dripping teapot invention would not need to be enabled over the entire scope of possible teapot materials. 

Conversely, to be sufficiently disclosed, the non-dripping teapot should be enabled over all possible spout lengths that provide the non-drip feature. If, for example, the spout length required for non-dripping was so long that a new type of material had to be invented to provide the appropriate stability, and this material was invented after the teapot invention priority date, the non-dripping teapot would not be enabled under Regeneron. 

Turning back to the case in question. MGI argued that the Illumina patent claims included features that were not enabled across the entire scope of the claim. Particularly, the claims included in their scope nucleotides that could be modified such that could not be used in the claimed sequencing method (para. 247). MGI also argued that the claims covered methods of sequencing with unspecified read-lengths, whilst the specification only included examples of short reads. Mr Justice Birss dismissed both of these embodiments as effectively analogous to chocolate teapots that did not relate to the essence of the invention (para 288). For example, whilst it was possible to conceive of impractical modified nucleotides covered by the scope of the claim, a skilled person would readily disregard these versions in favour of a functioning embodiment. MGI's sufficiency attack was thus dismissed, and Illumina's patents upheld. 

Final thoughts

In his decision on sufficiency, Mr Justice Birss thus makes a clear distinction between insufficiency due to undue burden, and insufficiency in the Regeneron sense. Mr Justice Birss stresses that a crucial part of a sufficiency analysis is determining which elements of a claim are relevant for Regeneron sufficiency. Crucially, the relevant features for Regeneron sufficiency are those relating to the "essence of the invention". 

The "essence of the invention" is a concept favoured by English judges which also turns up in the application of the doctrine of equivalents under Actavis (see Regen Lab v Estar [2019] EWHC 63 (Pat), IPKat). In the present case, identifying the essence of the invention seemed relatively straight-forward. However, as seen with the EPO's problem-solution prior art, identifying the problem-to-be-solved, and hence the technical contribution of a claim, is not always straightforward at the claim drafting stage when the inventor may be unaware of the closet prior art or the full potential of their invention. The decision in Regeneron was a reminder that the full contribution of a patent claim, for both validity and infringement, may be one that is decided years later, in the courts. The decision in Illumina confirms that such an analysis of sufficiency should at least be grounded in practicality and not a universe of infinite irrelevant embodiments. 

Further reading

Illumina v MGI Part 2: Has the UK lost its way on the doctrine of equivalents?