Do Lord Justice Arnold’s comments on Fibrogen v Akebia signify a change in approach to instruction of experts?


This article was first published in the May-June issue of the Patent Lawyer Magazine.

On 20 April 2020, Arnold LJ (sitting as a High Court judge) handed down a comprehensive 640 paragraph judgment in FibroGen v Akebia [2020] EWHC 866 (Pat). As well as dealing with a multitude of case-specific issues, the judgment also contains a number of more general practice points (and warnings) which are likely to be of interest to UK patent practitioners.

First, the judge re-iterated the warning he gave in Medimmune v Novartis [2011] EWHC 1669 (Pat) that lawyers should ensure their experts disclose their own relevant publications and, where appropriate, explain them in their report. The expert’s CV should also be included.

He also re-iterated the warning he gave in MSD v Shionogi [2016] EWHC 2989 (Pat) regarding personal attacks on witnesses during crossexamination, as in this case, counsel had allegedly continued to berate the witness for a page of the transcript regarding whether some papers referred to in his report had been supplied to him by his instructing lawyers, in circumstances where the expert had already provided a response, and without any basis to doubt his previous response.

It was, however, the judge’s comments on “sequential unmasking”, which grabbed this reader’s attention; the process of instructing anexpert in a particular order to minimize hindsight bias – first asking about the common general knowledge, next showing the expert the prior art and asking questions such as what steps would be obvious in light of that art, and only then showing them the patent in suit. This practice was first commended by Mr Justice Arnold (as he then was) in Medimmune v Novartis and is sometimes referred to as the “Medimmune approach”.

Practitioners will be familiar with the practical difficulties of the Medimmune approach. Discussing the common general knowledge and the prior art (and obvious developments therefrom), and only then the patent, and documenting the expert’s views at each stage, is a very time consuming and expensive process, especially where it is necessary to speak to a number of potential experts to find someone appropriate to support the case. The staged approach is also only as “sterile” as the explanation of the problem which the expert is being asked to solve from the common general knowledge; phrasing that question without suggesting an avenue which leads to the patent is in itself not easy.

There is, therefore, a lively debate as to the benefits and disadvantages of the Medimmune approach. In the absence of a clear judicial
consensus, and with large sums of money at stake and the arguments often finely balanced, the Medimmune approach has become generally accepted practice amongst patent practitioners, who simply cannot afford to take the risk that their expert’s evidence is ultimately given less weight as a result of the way in which they were instructed.

In this recent decision, Arnold LJ takes the opportunity to clarify his views on instruction of experts. In particular, he confirms there is no rule
that experts should be instructed using the Medimmune approach and acknowledges that there are often real practical problems in doing so. The question for practitioners is, though, do the comments go far enough to make it safe for parties to abandon the rather cumbersome exercise generally?


The Medimmune approach became generally accepted practice following Mr Justice Arnold’s comments in Medimmune v Novartis, handed down in July 2011. In that case, he said that instructing the expert in sequence was “… the correct way for those instructing [the expert] to proceed…” and commented that it was unfortunate that the expert’s report was not drafted in a manner which reflected the opinions he expressed at each stage of the assessment. These comments are the reason why many expert reports today are laid out in an orderwhich reflects the Medimmune approach.

In HTC v Gemalto [2013] EWHC 1876 (Pat) handed down in July 2013, Mr Justice Birss appeared to take a somewhat different view. He suggested the Medimmune approach was not a concrete rule, although it had been useful in the particular context of that case. What mattered was the reasons for the expert’s opinion and not the opinion itself, as the expert is not a representative of the skilled person and their report is not a test to see what the skilled person would do before being handed the patent (otherwise the parties may have to disclose the results of similar tests they had carried out with other experts, which would add enormous cost and complexity to patents cases). He also recognised that many lawyers are highly skilled in presenting information to an expert which, consciously or subconsciously, has the effect of leading the expert to the conclusion sought, which is capable of giving a false impression.

Whilst the court’s view in Gemalto provided helpful guidance that the Medimmune approach was not mandatory, the potentially devastating consequences of not following the Medimmune approach were highlighted in HTC v Nokia [2013] EWHC 3247 (Pat) – where HTC’s expert evidence was strongly criticised because of the way the expert was instructed. In paragraph [5] of the judgment, the judge noted:

“Furthermore, it became apparent that he had been instructed in a manner which was calculated to induce hindsight. Initially he was provided with the Patent and some prior art. To begin with, he focused on the Patent and wrote a critique of it which he accepted accentuated the negative over the positive. Only subsequently did he focus on the prior art, but even then he understood from his instructions that he was only to consider parts of Itakura. As a result, he only considered the document as a whole a couple of weeks before trial…”

It is perhaps unsurprising that HTC’s invalidity (and non-infringement) arguments were rejected by the judge in their entirety.

There are more examples in the case law of experts being criticized for using hindsight by virtue of the manner in which they are instructed – see for example Starsight Inc v Virgin Media [2014] EWHC 828 (Pat) at [131] and American Science v Rapiscan [2016] EWHC 756 (Pat) at [109] to [111]. By contrast, in Rovi v Virgin Media [2014] EWHC 1559 (Pat), Mann J did not consider the fact that Virgin’s expert had been shown the patent before the prior art to be a relevant criticism on the facts of the case, the judge observing that “I am not sure what he would have been able to contribute had he been shown the prior art with no real context against which to consider it”

Interested readers can review a report of a seminar hosted by LSE in March 20181 at which this issue was further debated. Of note was Arnold J’s opinion that the courts are highly dependent on high quality expert evidence and that parties should therefore want to ameliorate potential weaknesses such as hindsight bias as much as possible. Birss J suggested a note of caution however that if blinding is to be used, double blinding may be needed (i.e. the expert must not know which side they are instructed by), the judges may also need to be blinded, and the difference in skill between the expert and the skilled person may need to be considered.

Although it is clear from the case law that non-adherence to the Medimmune approach is not necessarily fatal, the fundamental importance of expert evidence in UK patent proceedings means that parties have felt to date that they are potentially taking a risk if they do not follow the dance steps when it might later be said that they could have done so.

The decision in FibroGen

In FibroGen, the defendants’ expert, Prof Haase, was accused of using hindsight in reaching his conclusion that the claims were obvious because he had been shown the PCT application for the patent before the prior art. The defendants justified their approach on the basis that obviousness was not in issue when Prof Haase was first approached and Prof Haase needed the PCT application in order to identify the skilled team to which the patent is addressed.

In addressing this criticism, Arnold LJ commented that the submission illustrates why it can be advantageous to instruct expert witnesses using the sequential unmasking approach, and re-iterated his view that it was desirable to minimise hindsight on the part of the expert where possible. However, he also confirmed that there was “no rule or principle hat experts must be instructed sequentially” and that “there are often real practical problems in doing so”. He then gave examples of such problems, including:

  • how does one identify the skilled person in order to establish the common general knowledge without showing the expert the patent?
  • how does the expert put his pre-existing knowledge out of his head when considering the prior art? What happens if new prior art is subsequently introduced?

He noted that even if an expert is instructed in sequence, it is often difficult for them to recall what they knew when, unless it is clearly documented.

In this case, Arnold LJ noted that because Prof Haase had prior knowledge of the subject matter of the invention, it would not have been possible to instruct him in a way which was free from hindsight. He therefore did not criticise the manner in which he was instructed but accepted that it should be taken into account in assessment of his evidence. Ultimately, the defendants’ obviousness attack failed although this did not appear to turn expressly on the manner in which Prof Haase had been instructed.

Where does this leave practitioners?

The FibroGen decision is helpful to practitioners in that it confirms the Medimmune approach is not a rule that must always be followed and recognises the inherent difficulties in the approach. However it still leaves open the question of whether all other things being equal, the court will prefer expert evidence which has been obtained following the Medimmune approach (or at least, be in a position to use such a disparity of approach as support for a finding leaning one way or the other). Practitioners are under increasing pressure from the courts to keep costs proportionate, and from clients to keep costs to a minimum. In circumstances where the stakes are high, the judge’s comments in FibroGen may perhaps still leave them with a difficult decision to resolve.

Matthew Raynor


Related Articles